presenteD to
£be Xibran?
of tbe
Tllntvereitp of Toronto
Digitized by the Internet Archive
in 2010 with funding from
University of Toronto
http://www.archive.org/details/p1iecindustriale13amer
Chem%^i^f^
Cfte Journal or fntmsitrtal
ano engineering CJemtsrrp
PUBLISHED BY
%f)t &mertean Cfjemteal g>octetp
VOLUME XIII, 1921
2£>oart> of CDitors
Editor: CHAS. H. HERTY
Assistant Editor: Lois W. Woodford
Advisory Board
H. E. Barnard J. W. Beckman A. D. Little A. V. H. Mory
Chas. L. Reese Geo. D. Rosengarten T. B. Wagner
EASTON. PA.
ESCHENBACH PRINTING COMPANY
1921
/
:°
L /3
v. I3>
nt .1
11
TTPiq c/oztrnal <yP
INDU
RIAL
%> ENGINEERING
CHEMISTRY
Published Monthly by The American Chemical Society
Advisory Board: H. E. Barnard
Chas. L. Reese
Bditobial Offices :
One Madison Avenue, Room 343
New York City
Telephone: Gramercy 0613-0614
Editor: CHAS. H. HERTY
Assistant Editor: Lois W. Woodford
J. W. Beckman A. D. Little A. V. H. Mory
Geo. D. Rosengarten T. B. Wagner
Advbrtisinc Department:
1 70 Metropolitan Tower
New York City
Telephone: Gramercy 3880
Vok
13
JANUARY 1, 1921
No.
CONTENTS
Editorials:
Officers for 192 1 2
Will the Senate Act? 2
When a Law Defeats Itself — Repeal It! 3
Amenities de Luxe 3
Are Your Folks on the List? 3
No Time for Dullness 4
Expansion of the News Service 4
Equitable Distribution 5
Notes 5
Chemical Industry and Trade of France. O.P.Hopkins. 6
Fuel Symposium:
Low-Temperature Carbonization and Its Application
to High Oxygen Coals. S. W. Parr and T. E.
Layng 14
Carbonization of Canadian Lignite. Edgar Stansfield 17
The Commercial Realization of the Low-Temperature
Carbonization of Coal. Harry A. Curtis 23
By-Product Coking. F. W. Sperr, Jr., and E. H. Bird 26
By-Product Coke, Anthracite, and Pittsburgh Coal as
Fuel for Heating Houses. Henry Kreisinger 31
Some Factors Affecting the Sulfur Content of Coke and
Gas in the Carbonization of Coal. Alfred R.
Powell 33
The Distribution of the Forms of Sulfur in the Coal
Bed. H. F. Yancey and Thomas Fraser 35
.Colloidal Fuels, Their Preparation and Properties.
S. E. Sheppard 37
Fuel Conservation, Present and Future. Horace C.
Porter 47
Gasoline Losses Due to Incomplete Combustion in
Motor Vehicles. A. C. Fieldner, A. A. Straub and
G. W. Jones 51
Enrichment of Artificial Gas with Natural Gas. James
B. Garner 58
The Charcoal Method of Gasoline Recovery. G. A.
Burrell, G. G. Oberfell and C. L. Voress 58
Original Papers:
Studies on the Nitrotoluenes. V — Binary Systems
of o-Nitrotoluene and Another Nitrotoluene. James
M. Bell, Edward B. Cordon, Fletcher H. Spry and
Woodford White 59
The Preparation and Analysis of a Cattle Food Con-
sisting of Hydrolyzed Sawdust. E. C. Sherrard and
G. W. Blanco 61
The Effect of Concentration of Chrome Liquor upon
the Adsorption of Its Constituents by Hide Sub
stance. Arthur W. Thomas and Margaret W.
Kelly 65
— The Action of Certain Organic Accelerators in the
Vulcanization of Rubber. II— G. D. Kratz, A. H.
Flower and B. J. Shapiro 67
Electric Oven for Rapid Moisture Tests. Guilford L.
Spencer 70
Addresses and Contributed Articles:
—The Chemistry of Vitamines. Atherton Seidell 72
— The Mechanism of Catalytic Processes. Hugh S.
Taylor 7 1
Industrial and Agricultltral Chemistry in the British
West Indies, with Some Account of the Work of
Sir Francis Watts, Imperial Commissioner of
Agriculture. C. A. Browne 78
Research Problems in Colloid Chemistry. Wilder D.
Bancroft 83
Scientific Societies:
Crop Protection Institute Discusses War on Boll-
Weevil; American Institute of Chemical Engineers;
Association of Official Agricultural Chemists; Cal-
endar of Meetings; Perkin Medal Award; Corpora-
tion Members of the American Chemical Society.. . . Sq
Notes and Correspondence:
Pure Phthalic Anhydride; Standardization of Indus-
trial Laboratory Apparatus; American Institute of
Baking, Research Fellowships 91
Washington Letter '<-'
Paris Letter 94
London Letter 94
Personal Notes 95
Government Publications 'n
Book Reviews 99
New Publications 102
Market Report 103
Subscription to non-members. $7.50; single copy, 75 cents, to members. 60 cents. Foreign postage, 75 cents, Canada, Cuba and Mexico excepted.
Entered as Second class Matter December 19, 1908. at the Post Office at Easton. Pa., under the Act of March 3, 1879.
Acceptance for mailing at special rate of postage provided for in Section 1103, Act of October 3, 1917, authorized July 13, 1918.
Subscriptions and claims for lost copies should be referred to Charles L Parsons. Secretary. 1709 G Street. N W . Washington. D C
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol.
i 3, No. i
LDITORIALS
OFFICERS FOR 192 1
The result of the ballot, of the Council for officers of
the American Chemical Society for the current year
is as follows:
President
Edgar Pahs Smith
Directors
George D. Rosengarten
Henry P. Talbot
Councilors
H. E. Howe
C. L. Alsberc
Allen Rogers
Lauder W. Jones
WILL THE SENATE ACT?
The Sixty-sixth Congress ends on March 4, 192 1.
( tne^of the three months available for legislation at
this final session has passed into history, and the dye
bill still remains on the calendar of unfinished business.
The question is being asked by all "Will the Senate
act?" We repeat again our conviction that it will.
Every argument hitherto presented in behalf of the
legislation stands to-day as forceful as ever. To
these must be added now the easily evident fact
that the failure to pass this legislation has
brought about a degree of demoralization which is
lamentable. Contemplated expansion of plants has
been postponed because of the uncertainty of the
future, research staffs are being contracted, a short-
sighted policy on the part of manufacturers, but true
nevertheless in many cases, and the chilling effect of
this demoralization is making itself felt in the ranks of
our chemists and students of chemistry.
Now comes a new factor into the situation. In ad-
dition to the large amounts of new capital being called
for by the German dye cartel, the life of that cartel
has been extended from the year 1966 to 2000, and
its dissolution at that time made more difficult by
requiring a four-fifths instead of a two-thirds ma-
jority to effect its dissolution. Not content with this
unification the segregation of the nitrogen-fixation
industry under the Haber process has been accom-
plished by the formation of an organization capitalized
at 500,000,000 marks, which organization is placed
undei the eontrol of the dye cartel. Regaining
mastery in the field of dyes is now not sufficient, am-
bition is leading on to a world control of nitrogenous
products. That is a threat which no nation can
ignore. There is no secret about the matter. The
facts have all been published.
With this situation existing, can the Senate
afford not to act? On what grounds could delay
be justified? Senator Thomas' nightmare of an
American dye trust was refuted sufficiently by the
declaration of the great mass of small producers of
dyes, read on the floor of the Senate, that they would
be the first to go under in the price war which would
follow the failure to enact adequate legislation; but
the Senator's dream looks like thirty cents when com-
pared with the steps already taken in Germany to
secure domination of the world's dye and nitrogen
supplies. The press report that this fixed-nitrogen
organization is contemplating the erection of plants
in the United States and Japan may be erroneous,
but already the market situation is being felt out.
The following circular letter is being distributed in
the trade. One of our dye concerns, the Peerless
Color Company, Inc., of Bound Brook. N. J., has
furnished us a copy.
C. B. Peters Co., Inc.
15 Maiden Lane
New York
Peerless Color Co., Inc.,
Bound Brook, X. J.
Gentlemen:
nitrite of soda
As previously advised you, we have for distribution ia this
country through American fiscal agents, that portion of Nitrite
of Soda, as produced by the Badische Anilin- & Soda-Fabrik
of Germany through their atmospheric nitrogen development,
which has been allotted for consumption in the United
States.
Naturally because of the existing business depression, there-
is very little activity, with the result that prices have bee« re-
duced considerably; in fact for spot material we can offer, sub-
ject to change, ton lots as low as 6c per lb. ex warehouse at
New York, and for larger quantities it might be possible to
shade this figure with a firm bid in hand, although the feeling
here is very strong that the bottom of the market has been
reached. We have on hand at the present time in New York
approximately 50 tons, and no further shipments will come into
this country until orders are placed for shipment from
abroad.
We have instructions from Germany to find out the prospects
of Nitrite of Soda consumption in the United States over the
year 192 1, and for this reason we are taking the liberty of ad-
dressing you to ask if you will kindly let us have your opinion
in this regard. If the market has actually reached its lowest
level, this might be a good time to consider requirement con-
tracts for the coming year and any suggestions that buyers have,
we shall be happy to cable abroad. The quality of our material
is as good as that produced in any part of the world, and we
shall be pleased to forward samples upon request.
Awaiting with interest your reply, we remain
Yours very truly,
C. B. Peters Co., Inc.,
cbp-th (Signed) C. B. Peters, President
To this request the Company responded:
Please be advised that we shall not, under any conditions,
cooperate with you in supplying the information wanted by
the Germans nor will we knowingly buy one pound of the sur-
plus German air-fixation products at 6c per pound or any other
price.
Reports from Washington indicate that the Moses-
Thomas combination intends to filibuster as strenu-
ously as ever. Under ordinary procedure they can
defeat the bill. The favorable majority in the Sen-
ate, however, can thwart these tactics by adopting
a closure rule limiting debate on the bill. This is an
action rarely resorted to by the Senate, but the un-
yielding and inexplicably bitter opposition of this
very small minority, on the one hand, and the future
welfare of this country as involved in this new com-
bination threat from abroad, on the other hand,
justify and demand the adoption of the closure.
Jan., iQ2i
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
WHEN A LAW DEFEATS ITSELF— REPEAL IT!
A state law which is directly contrary to the spirit
and intent of a federal statute should be repealed.
Such is the case with portions of Paragraphs 8 and 9
of Chapter 911 of the Laws of New York, which
became effective May 24, 1920, placing an excise tax
on the production and sale of "tax-free" alcohol.
The National Prohibition Act was avowedly framed
for the two-fold purpose of prohibiting the manu-
facture and sale of intoxicating liquors and en-
couraging the production of alcohol for industrial and
scientific purposes. Due and ample provision is made
for the production and distribution, under govern-
mental supervision, of tax-free non-beverage alcohol.
The plain purpose of the law is to remove any dis-
crimination against alcohol as a chemical reagent in
industry and in scientific research. In the face of
this plain declaration by Congress the New Y<Jrk
law levies a tax of $0.30 on each gallon of such alcohol,
and $250 on each place where it is sold. The tax-
free use contemplated by the federal statute is nullified
by the excise tax of the state law.
A law which defeats itself should be repealed. What
has happened since the enactment of this law? The
large distributors of industrial alcohol have moved
their warehouses across the river, from New York
into New Jersey. The ferry fare is cheaper than the
excise tax. Large manufacturers who could readily
add to existing stocks of alcohol have found, in view
of the tax, that it is not worth while to put in de-
alcoholizers, and this potential source of an important
chemical reagent is lost.
The manufacture of alcohol in New York State is
dead, the expected revenue from the excise tax is nil.
Common sense demands that it be repealed. Why
burden the courts with litigation testing its con-
stitutionality?
Minister expressed his hearty support of Mr. Hoshi's intention.
Mr. Hoshi, thus assured of the correctness of his proposal,
brought the matter to the notice of the German representative.
AMENITIES DE LUXE
The following interesting item appeared in the
English monthly supplement of The Yakitgo Shuho,
issue of November 7, 1920, published at Tokyo.
2,000,000 MARK CONTRIBUTION TO GERMANY
Mr. Hajime Hoshi, President of the Hoshi Pharmaceutical
Co., is to be congratulated on the admiration he has elicited
among the Germans as well as his countrymen for his contri-
bution of 2,000,000 mark to Germany for the cause of science.
Under date of September 26, Mr. Hoshi addressed a letter to
Dr. Solf, German Ambassador in Tokyo, in which he expressed
his wish to contribute 2,000,000 mark to the German Govern-
ment to be used for the cause of chemical and pharmaceutical
science in Germany. Mr. Hoshi further stated in his letter
that he has been an admirer of Germany especially in respect
of chemical and pharmaceutical science made in Japan and
that his contribution is intended to repay in some way the great
debt Japan owes to Germany.
On October 5, Dr. Solf, German Ambassador, sent a reply to
Mr. Hoshi in which he said that Mr. Hoshi's offer for the 2,000,000
mark contribution had been forwarded to the German Govern-
ment which gladly accepted the donation and promised that
the money would be used for the purpose as intended by the
donor. Dr. Solf expressed his belief that Mr. Hoshi's generous
gift will have the effect of encouraging scientific researches and
•of bringing Japan and Germany into closer relations.
It is understood that Mr. Hoshi before broaching his offer to
the German Ambassador consulted the views of Baron Goto
about his intended offer to Germany and the former Foreign
It is easy to imagine the smile of genuine delight
as Mr. Hoshi takes down his Christmas stocking and
finds it filled with the oranges, raisins and nuts of
"admiration he has elicited among the Germans as
well as his countrymen." We fear, however, that he
will find the nuts not up to market standard, per-
haps rancid, the nuts of the Japanese dye manu-
facturers, who we learn in another column of the
same publication are in dire straits because of the
present lamentable condition of their industry.
What is meant by "an admirer of Germany especi-
ally in respect of chemical and pharmaceutical science
made in Japan" we frankly cannot guess, but we are
confident that it is a bouquet of some kind of Japanese
wild flowers.
The well-remembered former Minister of Foreign
Affairs, Dr. Solf, "promised that the money would
be used for the purpose as intended by the donor"- —
a comforting assurance, doubtless, if one is disposed
to forget little things like scraps of paper. Dr. Solf
is confident that the gift "will have the effect of en-
couraging scientific researches." That's fine. Never
mind about the drop being lost in the ocean, it's good
to know that "scientific researches" are going to be
encouraged in Germany. And then, too, every little
bit of outside help for research makes that much more
of the present large dividends from the prosperous
German chemical organizations available for invest-
ment in the enormous capitalization increase now in
progress.
Mr. Hoshi, possibly for fear of wounding the sen-
sibilities of those he would encourage, was not going
to take any chances as to "the correctness of his pro-
posal," so he sought the advice of the former Japanese
Foreign Minister, Baron Goto. The Baron said, "Go
to it!" At least that is a brief way of expressing his
concurrence. Thereupon Mr. Hoshi proceeded to
encourage. All in all it was an auspicious and il-
luminative occasion, and serves the purpose, as Dr.
Solf says, of "bringing Japan and Germany into
closer relations."
Maybe the example set by Mr. Hoshi will be fol-
lowed by the Oxford professors, now that they have
received the condescending forgiveness of their brother-
savants (not brother-servants as erroneously printed
in our December issue).
ARE YOUR FOLKS ON THE LIST ?
Is the firm or corporation with which you are con-
nected a corporation member of the American Chemi-
cal Society? If not, it should be.
If you can't answer the question look in the list of
corporation members on page 91 of this issue. If
you agree with the affirmation, and if the name is
not in that list, get busy!
The power of suggestion is strong. Try it on your
president or general manager. He should know how
many organizations are supporting the Society
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
through corporation membership. Let him know
that the Society is not an organization for the mere
selfish interest of its individual-chemist-members, but
that it seeks to serve the nation by creating a sound
public appreciation of the value of chemistry in every
line of industrial endeavor; that its activities are di-
rected to utilizing every legitimate agency to increase
the efficiency of the American chemist; that the in-
terest shown by this corporation membership is retro-
flexive through the quickened spirit of fellow-mem-
bership; and that there are certain direct perquisites
accruing to corporation members. These are given
in Section 7 of the Constitution of the Society.
Section 7. Any firm, corporation or association interested
in the promotion of chemistry may by vote of the Council be
elected to membership in the Society and shall after election
be known as a corporation member. A corporation member
shall have all the privileges of membership except that of holding
office, shall be sent the titles of all papers to be presented before
any General Section or Division of the Society; may on appli-
cation, made in advance of publication to the Editor of the
Journal of Industrial and Engineering Chemistry, be furnished
with not more than five reprints of any paper announced for
publication, and shall have the privilege of being represented
in any meeting of the Society by a delegate appointed by the
firm, corporation or association. Such firm, corporation or
association shall pay annual membership dues of twenty-five
dollars.
If you fail on the first attempt, go at it again. See
to it that when the supplemental lists are published
the name of your firm or corporation is included.
Secretary Parsons will furnish the application blank,
or write him that the preliminary work has received
a favorable response and that it is up to him to finish
the job. He'll do it.
Here is another phase of the question. Without
solicitation the Arthur H. Thomas Company has
become a corporation member, and Mr. Thomas and
six members of his firm are individual members of
the Society. Can you beat it? If so, send us the
facts, we will gladly publish them.
NO TIME FOR DULLNESS
From time to time we have heard it complained
that members are not interested in the local sections,
that times are dull, and programs for meetings difficult
to arrange.
In view of the tremendous amount of work waiting
to be done, of the many possibilities for useful service,
such lamentations raise the question, "Is the real
function of the local section understood?" Frankly,
we think that if such dull times prevail the funda-
mental atmosphere must be one of desire to get some-
thing out of the local section rather than to put some-
thing into it. If once the spirit of service prevailed,
innumerable activities would suggest themselves where-
by good might be done in our neighborhoods, and
interest in local section activities be keenly aroused. .
When a man gives to something, he begins to take
interest in that something.
A fine illustration of the point we are trying to
bring out is afforded by the Milwaukee Section.
They have not been content to meet at regular intervals
and listen to distinguished lecturers either from within
or without their membership, but their progressive
officers have looked about for a way to serve the City
of Milwaukee. One of the first fruits was a request
from the Mayor of Milwaukee that the Local Section
appoint a committee to study critically reports on
Milwaukee's water supply, and to make any other
suggestions which would overcome present difficulties
with the. water supply. Chairman John Arthur
Wilson appointed a live committee, and the Mayor
is so pleased with the spirit in which the Section
responded to his request that he has "expressed the
wish that the Section will take an interest in all mu-
nicipal affairs where its opinion may help the city
officials to do the right thing."
The public library in Milwaukee was found to be
inadequately equipped with chemical journals. It
was felt that this was a much broader question than
the selfish interest of the chemists themselves, and
that by improving this situation the City of Milwaukee
would be benefited. In this connection, Chairman
Wilson writes:
An investigation of Milwaukee's industries revealed a need
for a very complete file of the world's chemical publications.
It seemed meet and right that any expense incurred in gathering
together such a file should be borne by the industries that
would profit by it. The Committee therefore started a drive
for a fund of ten thousand dollars, the interest on which is to
be spent perpetually for the purchase of chemical journals to
be placed at the disposal of the public at the Milwaukee Li-
brary. Each firm is asked to contribute no more than it feels
it will profit by the undertaking, so there is no begging or asking
for charity involved. For the best results, it was deemed ad-
visable that the fund and all journals purchased from it should
remain the property of the Milwaukee Section, which has
pledged itself to place the journals at the Public Library or any
other place it may choose such that access to them shall be
had by the public. The Milwaukee Public Library in turn
has agreed to take care of the journals and place them at the
disposal of the public so long as is desired and has further agreed
to be guided in the matter of purchasing chemical books and
in other matters pertaining to the chemist by the advice of the
local section.
*****
The response of all firms thus far approached has been so hearty
and sympathetic that there seems to be no doubt about the
ultimate raising of the full ten thousand dollars, which should
give the Committee a steady income in excess of five hundred
dollars a year to be spent only for chemical and closely allied
journals. Any portion of the income not needed for current
numbers will be spent in getting all back numbers of the more
important journals and in binding.
A fine illustration of how the chemist can serve his
neighbors! It is a safe prediction that the lines of
public work thus opened are only forerunners of many
others which will prove beneficial to the City of Mil-
waukee, and that dullness will never enter that pub-
lic-spirited and enterprising local section.
The problems in each locality doubtless differ, but
the principle of service is the same in all, and its re-
ward will be equally stimulative.
EXPANSION OF THE NEWS SERVICE
The sympathetic interest of the Directors in the
work of the A. C. S. News Service makes possible
its expansion during the coming year. The line of
expansion is definitely marked out and is a logical
outcome of developments during the past year. The
weekly bulletins and monthly clip sheet, "The Chemi-
cal Round Table," have been sent to about nine
Jan., 192 1
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
hundred of the leading daily newspapers. Through
the efforts of the Technical Director, Mr. John Walker
Harrington, an organization furnishing "boiler-plate"
matter to a large number of weekly newspapers be-
came interested, and made use of our bulletins in the
material which it distributed to several thousand
weekly papers. It is proposed now to enlist the
interest of all the associations which furnish plate
matter and to give to each a special service, thereby
hoping to reach all the weekly newspapers. By this
means the matter sent out by the News Service will
receive a largely increased circulation.
To carry on this work effectively, it is necessary
that we have the strong cooperation of the Program
Committees of the various sections. Remember this
is a news service and the matter to be sent out will be
determined largely by papers read and announce-
ments made before the various local sections. A
certain amount of time is required for the preparation
and distribution of bulletins, which should reach their
destination several days in advance of the release date
if we are to obtain the best results.
In the offices of the News Service a diary is being
kept of the meetings to be held by each local section,
and it is urged that those in charge of the programs
notify Mr. Harrington regarding the lecturer and his
subject as quickly as possible after the program for
each meeting is determined. Then if speakers will
furnish the News Service well in advance a copy of
the address, or at least a full abstract, the work of
preparing accurate bulletins will be greatly facili-
tated.
There is a wonderful opportunity this year to get
results far exceeding the fine results of the last two
years. To make the most of this opportunity we
must pull together, leaving to Mr. Harrington's judg-
ment the question of whether or not the material
adapts itself to newspaper use. If chemistry is to
take its proper place in a democracy such as our
nation is, it can only be accomplished through the
agency of sympathetic, well-informed public under-
standing throughout our citizenry.
EQUITABLE DISTRIBUTION
Year by year The Chemical Engineering Catalog
has grown in size and contents, apace with the growth
of the American chemical industry. It is a veritable
chemical exposition on paper. With each succeeding
year the errors and omissions of previous years have
been corrected. To the chemist or purchasing agent
in need of supplies it is a mine of information.
In the shaping of these volumes the compilers have
had the benefit of the advice of special representatives
of each of the national organizations of chemists.
The volumes thus become in part the property of all
chemists and accordingly have in the past been fur-
nished on request, without charge. But this policy
led to an unfortunate result. The presence of one
volume in a library or laboratory created the desire
for more; consequently there was frequent congestion
in the distribution, and the edition was soon exhausted.
For the late-comers the banquet was over because
of gluttony.
In the light of this experience a new policy has been
adopted this year. The volume is now mailed on
receipt of a leasing fee of $2.00. The charging of
this small amount should deter no one who really
needs it from receiving a copy of the Catalog; at the
same time it is hoped thereby to distribute the edition
fairly throughout the industry.
Congratulations to the publishers of the 1920 vol-
ume! May their power of useful service increase as
the years go by!
The French are contemplating the holding of an-
nual expositions of their chemical industries.
A British court has ruled favorably on the legality
of the appropriation of £100,000 by Brunner, Mond
& Co., Ltd., for the furtherance of research and scien-
tific education.
The organization of the Rochester meeting is taking
shape rapidly as a result of the energetic action of
the following chairmen of sub-committees:
Entertainment Committee: Chari.es F. Hutchinson
Transportation Committee: Charles W. Markus
Excursion Committee: William Earle
Finance Committee: Herbert Eisenhardt
Publicity Committee: Benjamin V. Bush
Hotels Committee: Harry LeB. Gray
Registration and Information Committee: Harry A. Carpenter
Program Committee: ErlS M. Billinos
When in New York City you happen to see each
morning on Fulton Street an erect man, with pure
white hair and clear eye, walking eastward carrying
a lunch box — look close, it is Dr. Charles F. Chandler
on his daily walk to work at the offices of the Chemical
Foundation. He didn't worry when December the
sixth reminded him incidentally that he was 84 years
of age.
Harking back to the days of the controversy over
the use of platinum for jewelry as against its conserva-
tion for munitions, it was interesting to read in the
November 6, 1920, issue of the Saturday Evening
Post the following quotation written in 1875 by the
late W. Stanley Jevons: "The appearance of platinum
being inferior to that of silver or gold, it is seldom or
never employed for purposes of ornaments."
If the idea in the opening paragraph of a letter just
received becomes a habit among our fellow chemists
we may be able to make this section of This Journal
both interesting and serviceable:
"Whenever matters affecting the status of the
American chemical industries or of the Chemical
Warfare Service come to my attention the signal
flashes through my mind 'Tell it to Herty.' "
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
CHLMICAL INDUSTRY AND TRADE OF FRANCL1
By O. P. Hopkins
1824 Belmont Road, Washington, D. C.
One result of the war has been the growth of a keen
desire on the part of French manufacturers to achieve
independence of the German chemical industry.
Along certain limited lines the French had made good
progress before hostilities began, but probably in no
country was German dominance in the markets for
chemicals so pronounced as it was in France, and it is
common knowledge that in no other country to-day
is the desire to be free of German dominance in any
line so freely expressed as it is there.
The war struck directly at the French chemical
industry, as many of the factories were in the Nord
and Nord-Est districts. The effect of the loss of these
factories on the chemical industry can be judged from
the following figures for the whole country:
Before the war
End of August 1914.
End of August 1917
Number of
Chemical
Establishments
... 1,583
894
... 1.410
Number of
Workmen
78,892
35,470
93.667
In brief, the number of factories and workmen
engaged in manufacturing chemicals was reduced by
half as a result of the German invasion, but within
three years the number of workmen so engaged was
about 19 per cent greater than normal.
The chief effort, of course, was directed to organizing
chemical plants for the production of munitions and
medicinal supplies for the army, and to direct this
effort there was organized the "Office des produits
chimiques et pharmaceutiques," under Professor
Bethal, the success of which has been demonstrated
by actual results. The obstacles faced by the French
at the outset can be appreciated if we consider what
our own plight would have been if half our chemical
industries had been taken from us within a week or so
of our entrance into the war.
As in other countries, there is now a desire to utilize
to the full in peace times the productive capacity
created during the war, but, as in other countries,
there is a growing realization that similar development
along exactly similar lines occurred in other countries,
and that much of the capacity so recently developed
will have to be adapted to other products or allowed to
stand idle. It is understood that this condition points
to spirited competition from the greatest industrial
nations, including England, the United States, and
Germany, and that the way to chemical independence
will be a difficult and trying one.
The chief development during the war occurred in
the production of heavy chemicals, statistics of which
are shown in the following table:
1 Facts and figures in this article are based upon publications of the
French government, upon the semi-official "French Year Book," upon the
German "Gluckauf," and upon published material issued by the United
States Bureau of Foreign and Domestic Commerce.
1919
1913
Productive
Production
Capacity
Metric Tons
Metric Tons
Sulfuric acid, ^8°
1.160,000
2,500,000
Sulfuric acid, 66°
58,000
1,200,000
6,000
300,000
Nitric acid
20.000
360,000
Sodium salts
625,000
800,000
Liquid chlorine
300
90.000
Bromine
500
Calcium carbide
32,000
200,000
Cyanamide
7,500
300,000
Ammonium salts.
75,000
200,000
Nitrate of lime
250.000
Natural phosphate
. 2,700,000
3,000.000
Superphosphates
1,965,000
2,500.000
Phosphorus
300
3.600
The foregoing figures do not cover the newly acquired
capacity for producing potash, which is discussed in the
section devoted to Alsace-Lorraine. The increased
capacity for producing nitrogen products, so noticeable
in these statistics, is referred to under the heading
"Fertilizers," and further comment will be found
under the heading "Heavy Chemicals."
Before the war France exported something like
$30,000,000 worth of chemicals, but the export trade
has been slow in recovering. On the other hand, the
import trade was brisk for a considerable period after
the war, as stocks of certain essentials needed re-
plenishing. During the last year French exports in
general have increased, and it is presumed that chemi-
cals have benefited along with other lines.
ALSACE-LORRAINE
By the return of Alsace-Lorraine, France has come
into possession of a district rich in agriculture, mineral
resources, and manufacturing industries. Of these
the most important in the building up of a greater
chemical industry are the minerals, the production of
which under German control in 1013 was as follows
(according to the ''Gluckauf'):
Number
Minerals
Establish-
Production
Metric Tons
21 ,135.554
3,795,932
8
76,672
6
49.584
1
6,354
The acquisition of the iron-ore resources of Lorraine
will make it possible for France to produce 40,000,000
tons of ore annually, and place her a good second after
the United States in this respect. Before the war she
was third, between Germany and England. The loss
of these deposits is a very serious matter for Germany,
as she formerly depended upon them for three-fourths
of the ore she needed. The manufacture of iron and
steel in the Lorraine district is very highly developed.
In 1913 France consumed 63,000,000 tons of coal, of
which 23,000,000 tons were imported. The bulk of
the domestic supply came from mines in the Nord and
Pas-de-Calais regions which were destroyed or damaged
during the war. The production of the Lorraine mines
was approximately 4,000,000 tons under German
control, and the production of the mines in that portion
Jan., k).m
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
of the Saare basin to be held by France until the
plebiscite 15 years hence was 12,000,000 tons. The
acquisition of this total of 16,000,000 tons will not
make France independent of other coal-producing
countries, especially not until the Nord and Pas-de-
Calais mines have been repaired, but French engineers
believe that the annexed fields can be developed to a
point that will insure eventual independence. The
future will depend upon French initiative and organiza-
tion.
The potash resources of the annexed territory are
estimated at from one and a half to two billion tons
of raw salt (say, 300,000 tons of K20), and it is con-
sidered possible that within a few years the annual
production will amount to 4,000,000 tons. The
present output is far from that, although it is nearly
four times what it was under German control. For
the last eight years the amount of crude salts mined
has been as follows:
Year
1913
1914
1915.
1916
1917
1918
1919
1920
Metric Ton
155,341
325,886
114,358
204.474
!20, 131
333.499
592,000
1,200,000'
1 Average daily production for August multiplied by 300.
Alsatian potash production before the war was
admittedly low, and the explanation generally offered
is that the mines were all new and that the output
was limited by the Kali-Syndicat to prevent over-
production. During the war, production fell off for a
number of reasons. One mine was bombed, and others
suffered from neglect and flooding. It is said that
some of the mines farthest from the front were badly
operated in an effort to speed up production.
Not all the damage done during the war has been
repaired, but it is evident that the mines in operation
are producing more effectively than they did under
German control. For the present they can be divided
into two groups, those under control of the Sequestra-
tion Office and those independent of that official
organization. There is considerable agitation for re-
moving all the mines from such control. Daily pro-
duction of all mines in August was 4000 tons of crude
salts, while the capacity was put at 8500 tons (7000
tons for the mines under sequestration and 1500 tons
for the others). It is calculated that with all the mines
in operation the production four years hence should
reach 14,000 tons a day. Perhaps a third of the
present production is going to the United States.
HEAVY CHEMICALS
France has been able to supply its own needs for
many of the heavy chemicals, as the table of imports
will prove. Before the war sulfuric acid was produced
to the extent of more than 1,000,000 tons, nitric acid
to the extent of about 20,000 tons, and hydrochloric
acid to the extent of some 130,000 tons. Com-
paratively small quantities were imported and ex-
ported. The war about doubled the capacity for
producing sulfuric acid, and the output of nitric and
hydrochloric acids was also greatly stimulated, so
that after the armistice there was an excess for export
with but few buyers, as a number of other countries
were in the same predicament. Soda products were
also manufactured to a sufficient extent to meet
domestic demands before the war, with a surplus for
export, and doubtless the same will be true as to potash
products as soon as the chemical industry has grown
up to the possibilities of the newly acquired Alsatian
resources.
■ FERTILIZERS
The war has opened the way to complete inde-
pendence for French agriculture so far as foreign
fertilizers are concerned. The need of nitric acid
in the manufacture of munitions led to a great develop-
ment of the nitrogen industry, just as it did in many
other countries, and efforts are now being concentrated
on keeping these new plants in operation on such
products as cyanamide and calcium nitrate. Cyan-
amide is now manufactured to the extent of more
than 100,000 tons annually, as contrasted with 7500
tons before the war, and French authorities have high
hopes of getting along without the 300,000 tons of
sodium nitrate formerly brought from Chile, although
they appreciate the fact that other countries have
ambitions along the same line, especially Germany
with its Haber process.
The acquisition of Alsace-Lorraine assures inde-
pendence of the Kali-Syndicat, and some export busi-'
ness in addition.
The production of superphosphates now amounts to
nearly 2,000,000 tons a year, which is sufficient to
meet the domestic demand. This industry operates
on phosphates from Morocco and Algeria.
COAL-TAR DYES
France is one of the half-dozen countries (i. e.,
France, England, Switzerland, Italy, Japan, and the
United States) avowedly seeking to establish dyestuff
industries that will make them independent of the
German manufacturers who formerly dominated the
world markets. In some respects the obstacles she
has to overcome are more serious than those con-
fronting the United States and England. The home
market is not extensive (imports of German dyes did
not exceed $3,000,000 before the war), and it requires
less in the way of staples and much more in the way of
specialties, since the product of the silk, wool, and
cotton industries consists largely of the most highly
finished fabrics. And the fact that so many other
countries are in the dye-making business will make it
difficult to find markets abroad for French dyes. On
the other hand, the value of a dye industry to the
national defense is more generally recognized and con-
ceded than in some other countries, notably the United
States, and the government has already armed itself
with the power to regulate the importation of German
dyes. (See the section headed "Government Assis-
tance.")
Authoritative figures on the present production of
artificial dyes are apparently not to be had and no
attempt will be made in this article to estimate the
output, but it is certain that no success comparable
to that of the American industry has been attained
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol.
No.
up to this time; in fact, American dyes and dyestuffs
have been marketed in Prance in fairly large quantities.
PERFUMERY AND COSMETICS
Among the highly finished luxury goods for which
France is famous are included perfumery and cosmetics,
by which are meant perfumes, essential oils, scented
soaps, grease paints, beauty creams, etc. The pro-
duction of these articles totaled in value some
$30,000,000 before the war, and they were exported to
all corners of the earth. This was an industry
naturally hit very hard by the war, but just as naturally
it made a very quick recovery as soon as the armistice
was signed and the period of luxury-buying set in.
It is the only important French chemical industry
that fared better in the export trade in 191 9 than in
IOIJ-
For some time before the war the French manu-
facturers of natural perfumes were somewhat worried
by the competition from German artificial scents, but
the French themselves are now manufacturing these
synthetic perfumes on an increasing scale, coincident
with the production of artificial dyes, and it seems
logical to assume that the long-established supremacy
in the natural products will assure the success of the
new industry. Not only have the new artificial scents
been favorably received, but considerable success has
been attained in blending the natural and artificial
products.
OILS AND SOAP
Marseille was a commanding figure in the vegetable-
oil and soap business before the war, the product of
its crushers amounting to some 1000 tons a day,
while the output of soap reached a very high figure.
Oil-bearing materials were brought to this port from
points in the Mediterranean and especially from the
Indian Ocean and the Far East by way of the Suez
Canal, and considerable quantities of more or less
crude oils were brought in for refining. The total
value of the products of the oil industries was
$86,000,000, of which Marseille was credited with
$70,000,000, Nice with $10,000,000, and Bordeaux with
less than $3,000,000.
The war interfered greatly with the importation of
oil-bearing materials, and a fat famine lasted until
long after the armistice. Even in 19 19 the imports
of oil-bearing materials were less than half what they
were in 1913. Peanuts, the principal raw material
crushed at Marseille, were imported to the extent of
nearly 500,000 tons in 1913, but in 1919 the total
quantity was only 225,000 tons. The falling off in
receipts of linseed and copra, the next most important
materials, is equally striking. Imports of oils in
1919 were much greater than in 1913, whereas the
exports dropped from about 58,000 tons to less than
8,000. Eventually Marseille will recover much of its
former business, but the development of the oil in-
dustries in England and the United States, to say
nothing of the tendency to crush near the source of
supply of the raw materials, are factors that are re-
ceiving serious consideration in France.
The production of common soap was affected by the
scarcity of fats during the war and is slow to return
to normal. Exports, which totaled nearly 78,000.000
lbs. in 1913, were 43 per cent below that figure in
1 91 9. In striking contrast to the decline in sales of
common soap is the increase in exports of scented soap
from a little over 3,000,000 lbs. in 1913 to nearly
7.000,000 lbs. in 1919.
GOVERNMENT ASSISTANCE
Protection by the government is a most important
factor in the development of a self-contained and
independent chemical industry in any country, or
of any branch of the chemical industry, and the chances
of ultimate success in the numerous countries that have
announced their intention of going their own way since
the war started can be appraised with some measure
of accuracy by a study of the steps taken to restrain
outside competition, especially German, until the home
industry can establish itself on a sound basis.
In France, as in the United States, England, Italy,
and Japan, there have been more or less whole-
hearted and intelligent efforts to foster a number of
chemical industries (coal-tar dyestuffs and medicinals
in particular) in the hope of ending the former German
monopoly, and the French government has to date
placed its reliance on high tariffs plus control of German
imports. There was a tariff on intermediates and
finished dyestuffs before the war, but it was un-
scientific in that the duty on the finished dyes was
much higher than that on the intermediates and was
the same for an intermediate that required little
finishing as for one that required a great deal of manu-
facturing to finish. The result was that the Germans
established finishing plants in France and defeated
both the revenue and protective objects of the tariff.
The new tariff is frankly protective and the rates
are not only higher but so adjusted that intermediates
requiring little labor to finish are only slightly lower
than the finished dyes, thus making it unlikely that
foreign manufacturers will be tempted to establish
mere "assembling" plants in France.
As against Swiss, British, and American competition
the tariff is at present the only protection afforded the
French dye-maker, and there is a disposition to complain
of the extent to which non-German foreign dyes have en-
tered the market. Against dyes of German origin there
is a licensing provision in addition to the tariff, al-
though the reparation allotments come in free of duty.
The decree upon which the French licensing program
is based may be continued indefinitely, differing in
that respect from our own war-time power to license
imports. In brief, the French dye-maker is ap-
parently assured of adequate protection against the
German dye industry, and thus better prepared for
eventualities than our own manufacturers.
THE MARKET FOR IMPORTED CHEMICALS
A study of the following compilation from official
French statistics shows how the wTar has affected the
French market for foreign chemicals, and incidentally
reveals the fact that the United States did not figure
prominently in the pre-war trade. Statistics are not
available to show the origin of 19 19 imports.
Tan., iQ2i
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
hcals and Allied
1913
Pounds
Products
1916
HKM1CALS:
Acetate of copper, <
Acetate of lead ....
Germany
United States
Acetone
Germany
United Kingdom
United States
Acids :
Acetic
Arsenious
Carbonic, liquid
Citric, crystallize*
Citric, liquid
440
452,160
354.940
Fori
Gallic, crystallized
Hydrochloric
Hydrofluoric
HvdroftuosilK-ic
Lactic
Nitric
Oleic, of animal origin.
Belgium
Spain
United States
Oxalic
Germany
United Kingdom ...
United States
Phosphoric
Stearic
Belgium
Netherlands
United Kingdom... .
" lited States.
4.441 .210
2.709.040
845,910
107.810
5.510
653,670
147,490
268.740
63,270
250,440
141 ,320
14,550
6,399.800
34.390
4,850
561.520
1 .822,560
5.851.730
4.786,450
6.170
429,900
57,760
34,170
210,320
272,930
42,990
12,790
2,544,130
195,550
403.220
336,430
1.878,340
1,303,590
447,760
118.170
87,080
500,670
125,660
2,661,420
1 ,276,920
1,214,750
13,890
,750
178,790
18,960
95,680
4,594,210
302,470
,114,010
1.619,960
Sulfuric 21,827,300 139,634,170
Belgii
Germany
Italy
United States
Tannic
Germany
United States
Tartaric
Germany
Italy
Alcohol, amyl
Alum, ammonia or potash.
Aluminium:
Chloride
Hydrate
Oxide, anhydrous
Sulfate
Ammonia
63,757,680
6,697,200
$100,736
Sulfate, refined
United Kingdom. .
Salts, other, crude . .
Salts, other, refined. .
Germany
Norway
United Kingdom .
Antimony oxides
Germany
United Kingdom
United States
Arsenic sulfide
Ashes, vegetable, and ly
Ashes, beet-root
Barium dioxide
Bromides
Bromine, liquid
Germany
United States
Calcium:
Borate
Carbide
Chloride
Sulfide and bisulfide..
Chemicals, n. e. s. :
With alcoholic base:
Taxed by weight. . .
Taxed by value. . . .
Other:
Taxed by weight.
1 , 105,180
624,350
373,460
32,630
246.920
3,530
728.410
5,730
337,970
614.650
479.500
457,020
118,830
1,896,190
229,060
934.540
513,680
203,050
67,240
130,950
2,200
536,820
616,850
7,929,950
412,700
20,720
169.750
169.750
6,291.710
8,157.680
24,910
50,050
653,890
3,530
1,855,190
7,500
220
183,420
303.350
8,928,060
8,878,450
37,071,170
53,247,580
30,368,230
1,661,180
246,250
236;770
9,480
660
333,560
436,290
218,700
156,530
1 ,980
' 1^980
30,860
Taxed by value $2,438,390
Chlorine, liquefied
Chloroform
United Kingdom
United States
Citrate of calcium
Italy
Cobalt :
Oxide, pure
Zaffer, siliceous oxide, vitrified
oxides, smalt, and azure. . .
Salts, n. e. s
Cocaine, crude
Germany
Copper:
Oxide
Sulfate 41 .856,550
Belgium 1,287,270
United Kingdom 40 , 373 , 730
United States
IUher, acetic and sulfuric 47.840
Fluorides 168.880
See also Fertilizers.
72,320
440
4111
5,730
245,150
2 . 650
2,430
2,430
191 ,140
35,594,500
6,250,320
12,790
198,420
$3,200,490
7,425,830
77,600
72,750
4.850
2,113.130
2.080,500
440
S9,r>(.6, l-.il
908,300
44,970
358,690
1 ,570,570
280,650
881 .630
137,570
60,410
65,700
25,350
3,054,950
5,730
160,060
87,520
3,776,520
$245,496
330 ',030
6.830
12,130
2,430
6,090,270
2.047,430
2,973,590
262,350
809,540
17,640
222,890
2,308.240
35,208,480
11 ,083,520
91 ,930
558,870
$5,887,27 2
164,910
46,740
660
3,090
3,970
Of Chemicals
Chemicals {Continued) :
Formaldehyde
Germany
United States
Formates
Glycerol
Netherlands
United Kingdom
United States
Iodides and iodoform
Iodine, crude or refined
United Kingdom
United States
Iron:
Lactate
Oxide
Sulfate
Sulfate of iron and copper.
Lactates, n. e. s
Lactarine (casein)
Carbonate
Belgium
Germany. .
United States
Chromate
Oxide
Allied Products (Continued)
1913 1916 1919
Pounds Pounds Pound?
3.090
I .045,870
367,950
252,210
50,050
50!650
1,320
3,232,600
6.685,670
13,230
28,880
55,340
8,463,240
5,712,340
1,296,080
Germany
Salts, n. e. s .
Magnesia, calcined
Magnesium:
Carbonate
Italy
L'nited State-,
Chloride
Germany
United States
Sulfate
Germany
British India
Mercuric sulfide:
In lumps, natural or artificial .
Pulverized (vermilion)
Germany
United States
Methanol
Canada
Germany
United States
Milk sugar (lactose)
Nicotine salts
Germany
United States
Phosphorus:
Red
White
Potassium:'
Acetate
Arsenate
Chlorate
Carbonate and crude potash. .
Belgium
Germany
Russia
Chromate of potassium and
sodium
Germany
United Kingdom
Nitrate
Oxalate
Permanganate
Germany
Switzerland
Prussiate
Sulfite, bisulfite
Pyrolignites of:
Calcium
Iron
Lead
Quinine, sulfate, and other salts.
Silver salts
Sodium:
Acetate
Arsenate
Bicarbonate
Carbonate:
Crude
Refined
Chlorates of sodium, barium,
Hydroxide (caustic soda)
United Kingdom
United States
Hyposulfite
Silicate of sodium and potas-
Sulfate...'.'
Sulfite, bisulfite
Tetraborate (borax) :
Crude
Chile
United States
Refined or semi-refined
Salts, n. e. s
Tartrates:
Cream of tartar
Crude tartar
Crystals of tartar
1 See also Fertilizers
98.990
1 ,765,230
505,740
1 ,113,540
324.080
76.500
1 ,330,480
753,750
294,320
6.149,960
6,092,640
5,002,500
1 ,715,630
2.026,710
306,000
33,290
46,960
22.490
12,130
440
19,840
7.720
16,256,220
1,328,720
11.105,900
2,750,240
6.438,980
3,581,370
2,421,310
157.410
108,470
506,840
459,220
11,900
45,420
288.140
395,510
15.650
22,710
486,780
18,740
338,190
1 ,089,960
880
108,250
31,970
74.740
2,650
139,330
46,740
880
1 .363,120
561 ,520
2,200
1,100
29,320
6,827,350 10.325,130
1,330,050
108,250 38,360
1,204,830 6,568.230
3.208,830
324,960
1,895,750
1,676.170
4,035,340
1 ,283!.'>io
2, 458^370
1 ,760
570, 810
355,380
3,300
8,197,000 8,056,790
6,797,070
6.605,050
11 ,900
143,080
,149,570
32,630
63.050
81 ,570
37,480
30,640
18,300
345,020
2,144,660
505.740
212,520
1,032,640
30.860
3, 1 22. S50
251,100
47.377,780
30.084,050
1 .060.640
28,961.240
131,400
168,210
2,870
328,270
14,297,860
1 .102,970
8,465,530
46,960
670,420
37,040
603,630
14,990
THE JOURNAL OF IXDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 15. No. r
op Chemicals and Allied Products (.Continues)
1,751,790
17,860
4. 190
Chemicals {Concluded):
Tartrates (.Concluded):
Wine lees 21 . -
Other 71 .21(1
Thorium and cerium salts 3, J30
Tin
Chlorides 2.030,450
Germany 1 .853,850
United States 171,740
Oxide 103,620
Uranium oxide 102 . 290
Belgium 9,040
Germany 91,490
United kingdom 1,100
Zinc:
Oxide 12,888.760
Germany 4,146,190
Netherlands 4,515,460
United States 3,864,670 1
Sulfate 421,080 141.080
Sulfide 219,800
Coal-Tar Products
Products obtained directly by
distillation of coal tar 190,097,800
Belgium 36.267,880
Germany 84 . 665 , 000
Spain 1,080,480 2.7'>4.14"
United Kingdom 65,167,540 114.600.460
United States 1,653,890 1" 02
Products derived from products
obtained by distillation of
coal tar
Germany
Switzerland
United Kingdom
United States
Dyes derived from coal tar:
Alizarin, artificial
Germany
United Kingdom
Picric acid
Germany
United States
Other dyes
Germany
Switzerland
United Kingdom
United States
Dyeing and Tanning Materials:
Extracts of woods, barks, nuts,
and berries used for dyeing :
Black or violet extracts
British America
United States
» Garancine
United States
Indigo, natural
British India
Orchil, prepared:
Dried
Moist, in paste
Red or yellow extracts
Extracts of woods, barks, nuts,
and berries used for tanning
Chestnut and other
British India
United States
Nutgalls and
Switzerland
Quebracho . . .
Argentina
Germany . . .
ExPLOsrvHs:
Dynamite
Spain
Fireworks
Gunpowder
United States
Fertilizers
9,000.370
145.730
440
6. 420. '160 14.955.940
37,628.840 (.5.206.100
S, 422. 240
7,790,620
237 . 220
150,350
8.600
660
660
1,263 250
410,280
87.080
271 ,390
15,430
220
48,060
.190.
774.040
601 .200
11.101.810
7,952,950
783,960
6.598,650 11.383.130
862.670
1 .388,030
4,335,390
8,215 (00 21. '89. 250
802.920
170.200
44,530
533,960
349,210
169.320
2,200
2,200
336.200
9.920
660
440
74,080
381,620
197,980
105.820
241.180
189,380
6.508.480
6,507,380
65.100.070
71.870
Ammonium sulfate, crude
Belgium
Germany
United Kingdom
Calcium nitrate and cyanamide.
Norway
Sweden
Switzerland
Fertilizers, chemical, n e. s
Belgium
Germany
United Kingdom
United States
Potash :
Muriate (chloride)
Germany
Italy
Sulfate
Germany
Metric Tons
20,696
4,110
8,237
8,123
10.010
9,378
232
400
223,217
28,860
157,107
31,709
430,120
190,220
6.610
13,654,550
12.668.860
Metric Tons Metric Tons
19.121
20,709
1 '- l
Slag, basic ....
Sodium nitrate
Chile
Superphosphate
Belgium
Tunis
United Kingdom
Medicinal Preparations
Distilled waters:
Alcoholic
1 Included
(')
322,115
322.014
100,822
83,983
828
540.700
540.694
4.122
1.498
156.169
118.255
12.956
6,130
Imports op Chemicals
Medicinal Prepns. (Concluded):
Distilled waters (.Concluded):
Nonalcoholic
Other, taxed by weight
United Kingdom
United States
Other, taxed by value
Oils, Fixed Vegetable:
■I and pulghere
Belgium
United Kingdom
United States
Coconut, touloucouna. illipe
palm nut
Belgium
Germany
United Kingdom
Colza
United Kingdom
Allied PRODUCTS (Continued)
1913 1916 1919
Pounds Pounds Pounds
52.030
152,780
76,940
21,380
$8,472
27,120
341 .060
61,950
80,690
$18,690
459,660 2.557,140 13.687,620
27,780
430,120 2.266.130
123.460
7,821.120
968,710
4.922,260
1 .726,220
59,520
For manufacture of soap . . .
Other
Cottonseed:
For manufacture of soap
edible fats
United Kingdom
United States
other
1 nited Kingdom
United States
I. HI
"Fertilizers, chemical,
n]1913.
China
United Kingdom
Mustard
Olive
Algeria
Greece
Italy
Spain
Tunis
Palm
China
West Africa, British
West Africa. French
Peanut:
For manufacture of soap or
edible fats
China
Japan
Other
Japan
United Kingdom
Rape
Sesame :
For manufacture of soap or
edible fats
Other
Soy-bean:
For manufacture of soap
Other
Other oils
Oils, Volatile:
Rose
Bulgaria
Germany
Switzerland
Rose geranium and vlang-ylang.
Algeria
Reunion
Other
British India
Germany
Indo-China
Italy
United Kingdom
Paints. Pigments. Varnishes:
Blacks:
For engraving
Ivory
Lampblack, Spanish black . . .
United States
Mineral, in lumps
Mineral, ground
Blue, Prussian
Carmines:
Common
Fine
Colors:
Ground in oil
In paste
I Xher
Green, mountain, Brunswick,
and other greens resulting
from a mixture of chromate of
lead and Prussian blue
Green. Schweinfurth, mitis green,
mountain blue and green ashes
Lithopone
Belgium
Germany
Netherlands
United States
Ultramarine
Varnishes:
Spirit
Turpentine, oil, or mixed
Germany
United Kingdom
United States
Zinc yellow, or chromate of zinc.
31,927,120
2,173.760
2,320.800
2,738,800
2,403,260
21 .604.420
34,729.820
2.761.950
1 .327,840
29,553,840
28.440
i5|430
401,460
14,550
145.280
337
179.770
61 .488
92,987
1,308,880
119.930
154,980
269,400
184.300
88.630
5,950
11.240
.411,630
.589.750
219,800
840.400
222.890
24 , 690
14.909.860
1,385,390
9.867,890
3,062,660
237,660
69.890
5,771 ,670
403,450
2,519,220
543.440
35.050
3.123.730 21. 484.05(1
28.880
10.446.600
2.983,260
7,347.790
9,087,890
3,622.860
5.449,380
4,671,820
1 ,269,420
1.732,170
2,581.170
6.166.990
1.757,300
2,933,030
9,359
1.745
6.710
4.938
25.040
638
24,311
50,870
15,444
115
1,052
18,601
15,302
70,437
2,718
720 5
840
210 3
790 4
780
560 3
540 5
020
260
iio 121
260
080
480
500 68
060 51
460 52
960
630
610
140.470
073,170
273.140
4,209.160
1 ,076.080
2,192.940
2,521 .427
1 .538,830
363.100
8,160
939.390
800,940
117.250
4.485,300
3 . 1 44 . 230
4.630
9.260
1 ,722.230
6.751,880
268.080
97
206,553
105,919
94.351
1 ,462,320
148.150
52,651
2.4J8!9io
185.630
258.600
223.770
53,1 10
115.080
.698.680
1.372.380
$87^056
35.490
276.680
243.170
5, 145.370
2.650
$280,622
10.140
3,581 ,410
304 . 240
99,870
910,950
622.800
21,600
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
Imports op Chemicals and Allied Products {Concluded)
Exports op Chemicals and Allied Products (.Continued)
Perfumhry and Cosmetics:
Alcoholic, gal
Nonalcoholic, lbs * . .
Oils, fixed, scented
Synthetic perfumes
Germany
Switzerland
Toilet soap:
Transparent
United Kingdom
United States
Other, scented
United Kingdom
United States
Miscellaneous Products:
Albumin
China
United Kingdom
Blacking
Candles:
Tallow
Wax and other. .
Italy
United Kingdom
United States
Dextrin
Gelatin, in powder, sheets, etc.
Italy
Switzerland
Germany
Glucose
United States
6.973
266,760
408
254,190
178,570
48,500
3,044,580
2,880,340
65,040
724,000
287,260
339,510
42,110
100,090
35,940
24,910
68,560
40,570
37,700
2,830
826,290
760,590
64.370
1,312,850
721,350
520,070
9,470
220
3,632,340
874,130
2,693,830
33,730
140,650
285,940
118,390
124,340
Glu
Germany
Switzerland
United Kingdom
Inks:
Drawing, in tablets.
Writing or printing.
Germany
United Kingdom.
United States
Isinglass
United Kingdom
United States
Paper and pulp:
Pulp, mechanical. . .
Germany
Canada
Norway
Russia
Sweden
Pulp, chemical
Austria-Hungary.
Germany
Norway
Sweden
Switzerland
United States
4.152,400
1,222,020
199,520
1 ,195,570
2,870
391,540
143,960
144,180
14,770
171,300
82,890
43,650
7,396,290
4,809,820
997,160
28,026,040
197
144,180
14,990
172,400 255,740
77,380
61,070
Metric Tons Metric Tons Metric Tons
259,449 213,209 161,168
6,396
4,504
115,923
15,380
116,342
205,500
26,536
42,716
31,830
88,803
4,606
2,711
Pounds
,480,400
942,030
177,690
36,252
107,293
4,636
844
Pounds
4,025,200
Paper, fancy
Germany
United Kingdom. . . .
United States 1,550.950
Paper, other 29,110,270 236,251,960
Germany 7,900,260
Norway 724,880 79,138,230
Sweden 3,149,300 95,147,530
United Kingdom 14,833,800
United States
Resin oil
Soap, common
United Kingdom
United States
Sugar (expressed in terms of
fined)
146,908,760
Russia
United Kingdom
United States
Turpentine, resins, rosin, pitch,
resin lumps, and other res-
inous products
Turpentine, spirits of
152.560
3,863,820
996,490
1,808,010
Metric Tons
108,062
Pounds
9,279.480
2,291,040
1,651 ,700
16.457,500
11,399,220
6,830 54,230
17,272.330 36,474,600
14,294,110
2.591,530
Metric Tons Metric Tons
543.126 568,867
Pounds Pounds
4,718,330 6,816,030
916.460
1,562.640
291.230
THE EXPORT TRADE
The details of the falling off in French exports of
chemicals in 1019 as compared with 1913 are shown
in the following table, which is based upon official
French statistics:
Exports of Chemicals
Allied Products
Chemicals:
Acetate of copper:
Crude 1,655,230
Russia 1,572,780
United States 15,210
Refined, powdered 721, 350
Crystallized 203 , 270
Acetate of lead 52,470
2,870
67,680
32,850
27,340
40,790
91 ,710
Chemicals (Continued):
Acetone
Acids:
Acetic
Arsenious
Boric
Belgium
Spain
United Kingdom
Carbonic, liquid
Citric, crystallized
Germany
United Kingdom
United States
Citric, liquid. ...'....
Formic
Gallic, crystallized
Hydrochloric
Hydrofluoric
Hydrofluosilicie
Lactic
Nitric
Belgium
Italy
Switzerland
Oleic, of animal origin . . . .
Belgium
Italy
Switzerland
Oxalic
Phosphoric
Stearic
Algeria
Italy
Switzerland
United States
Sulfuric
Tannic
Tartaric
Algeria
Germany
Spain
Switzerland
United Kingdom
United States
Alcohol, amyl
Belgium
United States
Alum of ammonia or potash .
Aluminium:
Chloride
Hydrate
Oxide, anhydrous
Norway
Switzerland
Sulfate
Argentina
Italy
Spain
12.350
395,730
2,472,700
4,749,190
1.072,100
190.260
2,284,870
655,210
896.840
248,240
98,550
488,540
16,498,710
Sulfate, refined
Algeria
Belgium
Free zones
Salts, other, crude
Salts, other, refined
Antimony oxides
Germany
United Kingdom
United States
Arsenic sulfide
United Kingdom
United States
Ashes, vegetable, and lye of . . . .
Ashes, beet root
Barium dioxide
Italy
United Kingdom
Bromides
Bromine, liquid
Calcium:
Borate
Carbide
Algeria
Morocco
Chloride
Belgium
Spain
United Kingdom
United States
Sulfite and bisulfite
Chemicals, n. e. s. :
With alcoholic base
United Kingdom
Other
Algeria
Belgium
Germany
United Kingdom
United States
Chlorine, liquefied
Chloroform
Citrate of calcium
Cobalt:
Oxide, pure
Zaffer, siliceous oxide, vitrified
oxides, smalt and azure. . . .
Salts, n. e. s
1 See also Fertilizers.
13,139,750
541,900
104,940
154,320
100.970
278,220
892,650
628,100
6,830
39,680
472,890
171,520
2,788,180
849,220
103,400
1,477,100
27,120
6,170
306,220
3.497,630
1 ,364,880
484.800
568,130
6,170
880
180,560
14,924,620
10,626,930
1,316.600
25,043,380
6,942.130
7,105,270
2,400,390
2,057,130
29.100
153,880
112,430
26,153,850
2,505,110
10,046,230
2,462,340
1.132.730
425,710
1,540
13,230
3,310
20,940
105,600
397,490
3,483.960
57,540
4.850
440
3,921,360
41 ,010
3,310
13,010
4,701 ,570
4,289,970
31 ,080
5,338,490
1 ,888.920
1,481 ,950
288,140
101 ,850
39,020
2,838,890
330,910
993,400
21,160
161,820
9,146,530
305,560
2,677,730
376,550
223,990
208,560
587,530
73,630
40,780
408 , 740
386,690
250,440
115,300
337,710
10,800
289,460
440
401,020
13,813,060
13,728,620
4,373,090
1,732,170
1,461,660
495,600
2,037,510
26,245,370
197,750
916,460
25.016,290
286,820
95,460
509,710
92,810
332,900
249,780
226,860
1 1 , 240
440
160,940
36,820
3,750
16,750
16,310
3,261 ,740
1,325,420
329,150
464,070
51,370
171,520
21,788,720
1,195,660
72,970
5,274,340
2,879.680
440
13,450
397,270
17,420
630,740
233,910
2,856,090
541 ,670
410,060
63,930
44,970
245,370
734,140
220
2,052,720
22,490
3,090
6,610
2,274,070
69 , 890
234,790
45,640
745,600
56,440
1,320
14,990
1 .078.720
1,749,590
464,510
1.155,220
191,800
945,120
3,530
71,430
501,770
344,360
16,530
128,970
7,280
3,272.100
24,690
1 ,813,300
246,920
7,270
17.420
379.200
8,668,570
65,920
1,345.040
2.200
1 ,496.060
13,309,750
186,510
314,380
74,520
255,080
516.760
6,830
1,050,280
440
17,420
274,250
18,li8i020
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
:ih:mic.\i i
i ocaine, ci ude
( oppi I
Oxide
Sulfate
Algeria
United Kingdom
Kther, acetic and sulfui i<
Fluorides
Formaldehyde
formates
Glycerol
Belgium
Italy ...
I tailed Kingdom
United States
Iodides and iodoform
Iodine, crude or refined
Iron:
Lactate
Oxide
Stdfate
Sulfate of iron and copper.
Lactates, n e. s
Lactarine (caseiiO
Germany
United Kingdom
United States
Lead:
Carbonate
L'hromate
Oxide
Salts, n. e. s
Magnesia, calcined
Magnesium:
Carbonate
Chloride
Sulfate
Mercuric sulfide:
In lumps, natural 01 artificial
Pulverized (vermilion)
Methanol
Milk sugar (lactose)
Nicotine salts
Phosphorus:
Red
Japan
Russia
Switzerland
White
Potassium:1
Acetate
Arsenate
Carbonate and crude potash.
United Kingdom
Chlorate
British India
Italy
Russia
Chromatc of pot a-
sodium
Nitrate
Oxalate
Permanganate
Prussiate
Sulfite and bisulfite
Pyrolignites of:
Calcium
Allied Products (Continued)
1913 1916 1919
Tounds Pounds Pounds
220
57,980
10,305,500
6.981 ,800
I .141,770
182,540
12,790
47,840
n.l
Ir
Quinine, sulfate and other
salts
Salts of thorium, cerium, etc. . . .
Silver salts
Sodium:
Acetate
Arsenate
Bicarbonate
Carbonate (soda, natural or
artificial
Crude
Algeria
Italy
Norway
ttcfined, not containing more
than 38 per cent of pure
l artionate
Refined, other
Algeria
Belgium
Netherlands
Sw itzerland
Chlorates of sodium, barium.
etc
Italy
Russia
Hydroxide (caustic soda)
Belgium
Netherlands
Switzerland
16.778.040
7 28.400
668,660
3.129.910
s. J13.180
5 2, 470
8 , 600
vS5
n
f>;
r-t
SMI 1
2
,430
14.582,250
6.721 ,670
2.263,040
3.917,170
683,870
10.140
1 . 152,350
12,130
97,440
20,060
110,890
406,750
24,030
215,390
198,200
1,100
8.380
35,050
7,984,100
4.196,720
3,021,650
2.037,510
622,140
128.310
24,690
24.470
I ,561.090
7.720
36,820
1 ,207,250
186,730
695.120
588.630
34.180
40,790
36,820
21,830
898,820
582,680
37,040
Sulfate
Belgium
Brazil
Italy
' See also Fertilizers.
7 , 125,550
174,398,200
3.656,140
96,006,010
30.238,380
18,645,370
1.787,950
284,840
506,180
29.622,850
14.977,100
6,277,660
5,887,660
144,840
213.850
653,230
53,302,700
23,658,460
16,760
9,347,600
5,437,040
2,673,550
47,400
27,. HO
64,150
97,660
8.273,070
1.399,270
6,743,500
9.480
20.7 20
9 !60
220
1 ,032,640
1,310,650
220
1,100
7,376,450
557,550
13,000
897,940
5,070
31 ,750
39,460
22,050
176.370
2,650
47.840
20.280
4,410
160,720
52,030
49,600
22,270
135,360
132.940
53,350
71 .210
24.470
21.660
4,410
421 ,080
51 ,150
592,820
113.980
61 .950
I !
5.7 14 7 '10
4.190
I .980
220.680
631 ,620
9.480
h 1711
.546,810
135.360
63,710
772.940
369,940
23,590
83,330
13,230
121,250
440
3,530
722,460
38,140
377,210
143.080
765,440
582,28(1
880
1 !
1 ,043.010
101.850
440
8,160
242,290
17,420
855,390
42,550
36.380
12,130
37,260
15,650
284,840
39,460
1.847,910
10,055.500 11,027,300
29,100
4,054,740
5,342,460
6,566,030 7,550,390
49.686,450 127. . 01 <"
5.734,880
132.720
41 .310,880
44.588,700 8.054 810
1 .801 .180
54 ,230
4,609,200 18.781.400
2,976'.o2o ;; '"
584,220
691.590 532.410
950 1.017.880
38,957,440 10
1 .964,760
9,378,460
3.802,310
3,287.530
Exports of Chemicals and Allied Products (Continued)
Chemicals (Concluded):
Sodium (Concluded) :
Sulfite and bisulfite
Tetraborate (borax):
Crude
Refined or semi-refined.
Belgium
Netherlands
Switzerland
United Kingdom
Salts, n. e. s
Tartrates:
Cream of tartar
Australia
United Kingdom
Crude tartar
United Kingdom
United States
Crystals of tartar
Wine lees
Other
Tin:
Chlorides
Oxide
Uranium oxide
Oxide
Ru
Spain
United Kingdom
United State*
Sulfate
Sulfide
Coal-Tar Products:
Products obtained directly by
distillation of coal tar
Products derived from products
obtained by distillation of
coal tar
Switzerland
Dyes derived from coal tar:
Alizarin, artificial
Picric acid
United Kingdom
Other dves
United Kingdom
United States
Indo-China
Dyeing and Tanning Materials
Extracts of woods, barks, nuts,
and berries used for dyeing
Black or violet
8,745.290
3,651,290
4,174,230
18,681.310
2.512,830
10.973,730
1,320
3.952,670
5,730
79,150
182,540
4,630
7,899.160
1 .180,790
380,520
1 .336.440
689,820
17,860
4.630
855,390
1 ,033,080
4.390,060
4,500,960
1,311,090
2,978,880
8,250,800
1 ,934,550
6,263,110
440
1,318,360
24,250
48.940
39,693.780
1,874,370
17,860
5,069,090
232.150
322.310
2,183,020
1 ..'07. 25(1
67,680
660
432,550
449.740
342.600
100.970
4,850
17,407,250 1.449,980 5.247.490
Chii
Germany
United Kingdon
United States. .
916.900
40.340
15.870
580,040
8 833.700
893.310
2,652.600
1 ,158,090
90,610
228,180
227,520
134.480
39,680
220
25,570
17,640
536] (80
1,100
71,210
164,020
Indigo, natural.
Indigo pastil, indigo bluing
Orchil, prepared:
Dried 28,000
Moist, in paste 25 ,350
Red or yellow 5.279,850
Italv
Spain 163,580
United Kingdom 1 ,589,310
United States 332,900
Extracts of woods, barks, tints,
and berries used for tanning:
Chestnut and other 207.113,030
Belgium 30,011.080
Germany 37,854.690
Indo-China 253,530
United Kingdom 97.079.440
Nutgalls and sumac 118.830
Quebracho 18.754,500
714,960
65,260
4,410
55,120
64,370
148,810
142,640
4,644,260
1,349,450
566,370
1,692,930
186,290
440
408.740
30.200
11 ,460
49.160
1 .769,210
29.200,440 15 ,.i ,, |0
Belgium.
United Kingdo
Algeria
Explosives:
Dvnamite
Algeria
2,184,780
5.958,430
30,420
793,440
26,012,780
3,090
471.7')" 26.273.150
Ru
Fireworks
Gunpowder
Algeria
Italy
Russia
Fertilizers:
Ammonium sulfate, crude
Calcium nitrate and cyanamide.
Fertilizers, chemical, n e s
Algeria
Belgium
Germany
Italy
United States
Potash:
Muriate (chloride)
Sulfate
Slag, basic
Sodium nitrate (Chile saltpeter)
Superphosphate
Algeria
Belgium
Italy
Portugal
Spain
Switzerland
430,780
.107.410
.525.380
220,020
Metric Tons
1,036
839
403 , 296
7,929
135,790
219,805
22,982
1 ,000
127
730
0)
5,268
145. 226
9.692
30,212
20.974
11.815
57,389
5.521
270.510
12,994.710
470,11(1
12,105.800
71 ,430
23,142,800
517.870
16,901.080
4,465,460
Metric Tons
1 ,328
5,511
3,078
2,571
! Included under "Fertilizers, chemical.
4,101
11,792
12,363
526
176
530
1,550
5.151
1.538
( 1913.
45.358
538
6,209
Jan., ig2r
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
Medicinal Prefab
Distilled waters:
, Alcoholic
United States'
Other compound-.
Argentina
Brazil
Cuba
Mexico
Spain
United Kingdom
United States
Oils, Fixed Vegetable:
Castor and pulghere
Allied Prod
1913
Gallons
97,798
14,028
52,783
Pounds
950
200
640
160
760
450
570
050
640
440
280
2,270
1,877
1,083
1 ,065
239
765
479
7,177
United Kingdom
Coconut, touloucouna, illipe,
palm nut
Italy
Switzerland
United Kingdom
United States
Colza
United Kingdom
United Stales
Cottonseed
Linseed
Algeria
Switzerland
Tunis
Mustard
Niger
Olh
Belgium
United Kingdom
United States
Palm
Italy
Switzerland
United Kingdom
United States .
Peanut
Algeria
Italy
Switzerland
United Kingdom.
United States
Poppyseed:
Black
White
Rape
Sesame
Algeria
Switzerland .....
United States
Soy-bean
Other oils
Switzerland
United Kingdom.
United States
Oils, Volatile:
1,035
21.643
2,676
3,819
3,046
6,794
4,041
582
897
2,035
5,783
1,514
13,027
2,672
1 ,270
2.292
2,429
763
420
414
53,427
18,511
6,799
4,769
4,356
5,269
495
3,695
255
4.347
906
247
Ro
Switzerland
United States
Rose geranium and ylang-ylang.
Germany
Spain
Switzerland
United States
Other
Germany
United Kingdom
United States
Paints, Pigments. Varnishes:
Blacks:
For engraving
Ivory
Lampblack
Belgium
Germany
Italy
Mineral, in lumps
Mineral, ground
Blue, Prussian
1,923
645
Tun
United States.
United Kingdo
Carmines:
13,230
1,855,850
433,650
806,230
206,570
279,330
301,150
220,240
23,370
37,260
British India
Colors:
Ground in oil
Algeria
Belgium
United Kingdom
United States
In paste
Other
Green, mountain, Brunswick,
and greens resulting from a
mixture of chromate of lead
and'jPrussian blue
13,890
660
5.950
4,850
924,180
1,174,180
402,790
;cts {Continued)
1916 1919
Gallons Gallons
33,154
4,993
Pounds
1
,110,690
181 ,000
311. 290
9
708,270
1
,471 .800
1
235,250
1
.653,250
432,770
327,160
359,570
1
,148,730
158,730
168,4 (0
11
,436,700
1
740,350
4
678,210
487.880
7.
.373,270
1
251,790
653.670
275.140
6
043,970
(,411,01111
3
461 .260
352,740
220
4
347,960
119,710
335,760
762,780
6
310,950
790,170
1
040,140
87.300
1
869 , 740
»s
Kofi. 8x0
7
603,960
150,800
9
265,370
1
531,990
1
977,110
27,340
321,870
4
518,380
1
826.530
1
801 ,190
49,600
18,080
709,890
| !9,630
135,140
103,840
12,965
3,693
133,294
3,580
18,667
61,600
1
124,800
229,500
255,520
440
9 , 260
558,650
192 680
65 , 260
195, 110
178.350
21 ,645.860
3,102,560
2.659.880
953 ,060
1 .162.060
3,090
61 .730
418.660
5 ,950
1 I 5 ,960
200,1.10
,870 3,160,550 3,146,220
110,230
84,220 24,030
794,760 419.1011 284,620
1,489,220 733.480 714,300
Paints. Etc {Concluded)
Green. Schweinfurth, 11
green, mountain blue
Lithopone
Ultramarine. . . .
Algeria
Egypt
LTnited Kingdom
United States
Varnishes:
Spirit
Turpentine, oil, or mixed.
Belgium
Italy
Spain
United Kingdom
Zinc yellow, or chromate of z
Perfumery and Cosmetics
Alcoholic
Argentina
Belgium
United Kingdom
United States
i.lied Products {Concluded)
1913 1916 1919
Pounds Pounds Pounds
74,960
225,750
3.784,670
496,040
767,210
176.370
Nonalcoholic
Argentina
Brazil
Belgium
United Kingdom.
United States
Oils, fixed, scented
Synthetic perfumes
LTnited Kingdom
United States
Toilet soap:
Transparent
United States
Other, scented
Algeria
British India
Indo-China
United Kingdom
United Slates
Miscellaneous Products:
Albumin
Germany
United States
Switzerland
Blacking
Belgium
Italy
United States
Candles:
Tallow
Wax and other
Algeria
Madagascar
Dextrin
Gelatin, in powder, she* ts eti
United Kingdom
United Stales
Glucose
238,980
3,402,610
686.300
629,420
249,560
326,940
1 ,100
Gallons
448,376
69,979
35,504
56,849
32,467
Pounds
5,333.860
286,160
112.440
615,970
1 .510,600
884,050
31,182
32,410
7,500
3,310
98 , 5 50
20,940
3,072,800
251 ,770
23,370
816,150
393.080
309,310
364,200
162,700
42,990
18,960
1,691 ,600
238,980
235,010
26,010
Glu
Belgium
Germany
United Kingdom
United States
Inks:
Drawing, in tablets.
Writing or printing .
Belgium
Brazil
Italy
United Kingdom .
Isinglass
United Kingdom
United States
Paper and pulp:
Pulp, mechanical
Pulp, chemical
Paper, fancy
United Kingdom .
United States ....
Paper, other
Algeria
Egypt
United Kingdom.
" lited States
188
6,772
5,707
252
306
1,016
624
118
347
16,605
3,836
955
6,454
912
720
160
540
210
220
330
130
830
450
000
700
700
250
270
1 oil.
Re
Soap, common
Algeria
Italy
Switzerland. . .
Tunis
United Kingdo
United States.
Turpentine. resin> rosin, pitch,
resin lumps and other indig
enous resinous products. . .
Switzerland
United Kingdom
Turpentine, spirits of
Italy
Switzerland
2,420
4,376,610
538,810
308 , 200
321,430
756,190
260,370
3,300
19.840
Metric Tons
59
594
Pounds
3,924,230
' 1,345,040
76,940
90,756,590
29,489,690
7,637,910
9,288.510
6.675,150
58,420
77,568,530
28.784,210
8,830,390
3,340,220
3,913,420
4,042,610
1 ,546,760
Metric Tons
199.115
Pounds
1 ,153.900
59.300
36,160
.182,590
345,020
677,920
324,520
31,330
393,080
283,960
110,230
3.310
Gallons
345.703
47,710
34|423
48.952
5,394,930
369,270
235,890
1 ,132 ',070
1,630,320
4,528
170,420
41,890
84,440
52,470
5,070
2,065,070
568,570
172,620
160,060
140,880
55,120
400,580
352.740
2.251,140
58,860
1 ,005,090
22,270
168,650
5.651,770
4,947,830
93,920
253,750
542,340
231,050
43,210
230,600
5,560,060
8,1
64, [50
3,970
Gallons
421 ,627
55.780
6. 944! 341 1
109,350
5,358,330
4,035,340
173.720
7,107,920
1,569,690
577,830
Metric Tons
5.070
2,644,890
23, L50
261 .470
309.970
379,200
404,550
74,520
139,330
Metric Tons
6 15
117 25
Pounds Pounds
1,533,100 1,546,540
483,690
87,080
66,659,840 43.470,300
23,613,710
771,400
5,724,740
10,813,670
128,090 65,480
53,463,410 44,304,090
28. ITS. 200
4,567,100
2,626,810
2,718,300
1,227,750
771,180
Metric Tons Metric Tons
94,486
Pounds
454,590
78.851
Pounds
384,260
90,159.570 67,470.700 114,201,640
1,149,710 8,688,420
19,707,340 34,152,900
21,525,270 6,065,800 14,959,240
3,653,060 1,301,610
,1,244,760 2,091.970
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. i
FULL SYMPOSIUM
Papers presented before the Div
of Industrial and Engineering Chemistry at the 60th Meeting of the American Chemical Society. Chii
September 6 to 10, 1920.
LOW-TEMPERATURE CARBONIZATION AND ITS APPLI-
CATION TO HIGH OXYGEN COALS
By S. W. Parr and T. E. Layng
University ov Illinois, Urbana, Illinois
The low-temperature carbonization of coal is ordi-
narily understood to mean its destructive distilla-
tion at temperatures not in excess of 750° or 800°
C.
COMPARISONS WITH HIGH-TEMPERATURE CARBONIZATION
Certain features which accompany this particular
condition may be briefly enumerated as follows:
The demarkation of temperatures indicated by
750° to 8oo° C. is not arbitrarily chosen, but seems to
be a natural dividing line between the decomposition
processes which liberate heavy products which are
largely condensable and those reactions which deliver
light or noncondensable compounds. Another method
of stating the case would be to say that below 750°
the volatile products are tars or oils and some fixed
gases, while above 750° the volatile products are gases
only.
Again, under low-temperature conditions the vola-
tile constituents are largely the initial products of
decomposition, as set free by the various components
of the coal, and in the main they are not subject to
any great modification by secondary processes of
decomposition. By this it is not intended to affirm
that no secondary reactions occur. By their very
nature, these volatile products are susceptible to
change, but these changes are more in the nature of
interactions or reactions among themselves or with
the decomposing constituents; whereas, under high-
temperature conditions, there proceeds a very positive
breaking down of these easily decomposable compounds.
In other words, the high-temperature process accen-
tuates the matter of secondary decomposition so that
the ultimate products bear little relation to the char-
acter of the substances that first result from the
destructive distillation of the coal.
yields — This contrast in products leads to the
next statement as to yields. A bituminous coal,
which under the ordinary high-temperature process
yields 10 gal. per ton of condensable material, will,
where these secondary decompositions are lacking,
yield from 20 to 25 gal. per ton. Indeed, certain types
of coal have been found where the condensable prod-
ucts are in excess of 30 gal. per ton.
CHARACTER OF LOW-TEMPERATURE PRODUCTS — -Other
interesting features relate to the character of the
compounds that are discharged under the low-tempera-
ture range. No information along this line can be
gained from a study of high-temperature products,
because their character has been quite altered or
obscured by the secondary decomposition resulting
from the passage of the initial volatile constituents
over or through the highly heated passageways or
masses of coke. As a matter of fact; it is only by
a study of the products as they are discharged at
successive temperature stages that we can arrive at
any safe conclusions as to the character of the initial
products of decomposition. It will not be strange,
therefore, if we have to modify to a considerable
extent our present conception of the decomposition
procedure.
Briefly stated, we shall find the order to be: water,
carbon dioxide, and methane, with respective tempera-
ture ranges of approximately 2500 to 3000, 300* to
3500, and 350° to 4000 C. At the latter stage, there
begins also the discharge of ethane and heavier hydro-
carbons, with the beginning also of condensable products
in which the sulfur and oxygen compounds predomi-
nate. The latter show themselves in the form of tar
acids. The chief feature concerning the sulfur is
that the part which is in organic combination in the
coal is quickly discharged. A range of temperature,
however, seems to be attained where there is sub-
stantially no sulfur decomposition, as shown by an
almost total absence of this constituent in the gases.
However, at higher temperatures where decomposition
of the iron pyrites occurs, the volatile sulfur compounds
appear, largely in combination with the tar or oil
constituents.
This substantial absence of secondary decomposition
accounts for a number of characteristic variations
in the by-products. For example, the tars are thin
and light, having a consistency much more resembling
oils. They have a specific gravity so nearly approach-
ing unity that, the separation of water from the oil
is difficult. The tars contain practically no free car-
bon. The gas yield per pound is less, being from 60 to
80 per cent of the volume obtained by high-temperature
processes, and both gas and tar are free from naph-
thalene.
These differences are such as one would naturally
expect as a result of the presence or absence of secon-
dary decompositions. The argument in favor of the
tars is that, in addition to their much higher yield,
it would be better to carry out the possible decomposi-
tions upon them as a distinct process under exact
control and for the production of specific substances,
rather than to submit them to the more or less uncer-
tain and haphazard reactions which result from the
high-temperature decompositions.
Another method of stating the important feature
of oil or tar yield is from the viewpoint of our rapidly
vanishing petroleum supplies. If, for example, a
Scotch shale with a yield of 20 or 25 gal. of oil per ton
and no by-products of value is a workable proposition,
why may we not look with favor upon a bituminous
coal having a potential yield of liquid fuel of 20 or 30
gal. per ton and a by-product in the way of a smokeless
solid fuel of even greater value than the oil?
Jan., 1021
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
IS
COKING OF HIGH OXYGEN COALS
Since the most of our high volatile coals are, as a
matter of fact, also high oxygen coals, the question
at once arises as to the possibility of producing a mar-
ketable coke from high oxygen coals. Concerning
the coking of such coals, we shall doubtless be obliged
to recast to a certain extent our theories concerning
the chemistry of coal carbonization.
theoretical considerations^ — -In a general way,
it has been held that a coal with an oxygen content
above a certain amount, for example, an oxygen-
hydrogen ratio much in excess of 50-50, or, say, 6 per cent
of oxygen to 5 per cent of hydrogen, should be classed as
a noncoking coal. This would seem a harsh decree for
Illinois coals, which exceed this oxygen ratio by almost
50 per cent; especially since the reserve tonnage of such
coals within the boundary of Illinois exceeds the
reserve tonnage of any other state in the Union,
Pennsylvania and West Virginia not excepted. Now
the fact that a low oxygen content is characteristic
of the coals which make good coke by methods now
in use may be a coincidence and not a cause. At
least, there has never been any very good explanation
of why a high oxygen content should result in poor
coke. We are positive in this connection only of one
thing, namely, that we have found no explanation
which we can guarantee as satisfactory in all cases, or
in all respects. But, experiments have proceeded to
a point where a few fundamental propositions are
seemingly established. For example, that part of
the coal which is "phenol-soluble"1 has a definite
melting point, and this material in its final decomposi-
tion furnishes the binder for the production of coke.
It is largely composed, however, of highly unsaturated
compounds, and these, if allowed to come in contact
with certain decomposition products of the fully
oxygenated type, unite with the same to form com-
pounds having totally different characteristics, chief
among which is the absence of any melting point,
and consequently the absence of the coking property.
Let us go a step further in this illustration. A coal
which is finely divided and which has been exposed
to the air for sometime will have lost its coking property,
even though the coal be of the so-called coking type.
Now, if our reasoning is correct, such a coal might
be so handled in the coking process as to eliminate
those oxygen compounds in such a manner as to
avoid the disastrous reactions with the active coking
constituents. Experimental evidence is in hand show-
ing this can be done. The same reasoning, of course,
will and does hold true for the coals with a high normal
oxygen content. They may be dealt with in such a
manner as to produce a very weak and indifferent
coke, as seen in the ordinary gas-house product, or
under other conditions where deleterious interactions
are avoided, a coke of altogether different texture and
density may be the result.
Further, these considerations are not inconsistent
with the theories now being developed by Doctor
Thiessen as to the composition of coal. He seems to
1 Pan- and Olin, University of Illinois Engineering Experiment Station,
show that the phenol-soluble portion is the degrada-
tion product, through geological processes, of cellulosic
material; and not, as Lewes would have us believe,
of resinic bodies. From this standpoint, we should
say, then, that this material which constitutes the
true coking substance has a marked tendency towards
a reversion of type. This may show itself either in
the interaction which occurs during the destructive
distillation process or more readily in the effect of
weathering. A striking illustration of the effect of
weathering is occasionally found in the case of Illinois
coals, where the outcrop shows a marked reversion
of type to the extent that it has every characteristic
of a lignite, whereas the coal from the working face,
completely removed from weathering effects, shows
no such reversion.
temperature control — -Thus far this discussion
has dealt only with some of the theories underlying
the carbonization of high oxygen coals. The methods
which suggest themselves for securing the conditions
indicated involve a procedure whereby the changes
may be brought about in stages or what may fairly
well be designated as fractional decompositions. Such
a method implies an observance of temperature con-
trol, quite unknown and quite impossible under the
ordinary high-temperature conditions. This matter
of temperature control involves the entire question
of successfully carrying out any sort of a low-tempera-
ture program. Indeed, it is of such paramount im-
portance, and in all of its bearings upon the situation
involves so many factors, that its proper discussion
should' be reserved for a separate consideration. How-
ever, brief reference is made here for the purpose of
indicating that the preceding discussion is not purely
academic and theoretical, with no hope of possible
attainment in practice, but, as a matter of fact, may
be found the most logical procedure even under indus-
trial conditions.
The first question we meet is this: Can we carry
heat to the center of a nonconducting mass by con-
ductivity methods alone, without doing violence to
all ideas of temperature control? If we look to the
modern by-product oven for an answer, we shall be
obliged to say at once, "No." In this practice, for
the temperature at the center of a coal mass of 18-in.
cross-section to reach the beginning of the carboniza-
tion stage requires at least 14 out of the total of 18
hrs.; and even this is accomplished only by main-
taining a surrounding temperature of 10000 as an
impelling force against the nonconductivity conditions
prevailing. Obviously the low-temperature idea in
any of its bearings is incompatible with such procedure.
A number of methods have been proposed for meeting
this condition of nonconductivity without the use of
excessive temperature. The most frequent is the
application of temperatures within the prescribed
limit to a mass of coal so narrow in its cross-section
that the penetration of heat from the two sides would
be sufficiently uniform and rapid to meet the require-
ments so far as ultimate temperature throughout the
mass is concerned. The same idea is involved in any
briquetting process with subsequent application of
i6
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
heat to the briquets, the factor involved being the cross-
section of the individual briquet masses.
In the process as we have been developing it, utiliza-
tion has been made of the ability of the coal under
proper conditions to supply its own heat, which may
thus be made to proceed autogenously throughout
the mass without reference to its size or cross-section,
and without the application of any external heat in
excess of the prescribed maximum for the theoretical
conditions involved in the low-temperature idea.
Fortunately, these reactions which are responsible
for what is well recognized as the exothermic behavior
of coal in the process of carbonization occur well within
the prescribed limits. As a matter of fact, they are
most in evidence at temperatures of approximately
300° to 400°. Up to date our experiments have not
involved cross sections of coal greater than 16 in.
character of coke obtained — The appearance
of the material produced under these conditions
is strikingly characteristic. It is uniform in texture,
without any zoning evidence of progressive stages in
heat transmission, dense, and of good strength, and
without any of the fingering effect characteristic of
the high-temperature method. The volatile matter
retained under these conditions may vary from 5 to
15 per cent, depending on the coal and the ultimate
temperature attained. It contains no condensable
hydrocarbons, and if discharged by application of
further heat would appear almost entirely as hydrogen
and methane. As would naturally be expected where
an autogenous generation of heat is involved, the
time element for bringing about the carbonization is
greatly reduced, the average time being from 3 to 4
hrs. Experiments involving the exact measurement
of the amount of heat available from different coals,
the conditions for its greatest development, and the
limits as to mass wherein it may be made practically
operative are still matters of experimental research.
DISCUSSION
Mr. J. D. Davis: I should like to question Professor Parr in re-
gard to the temperature at which naphthalene products begin
to show carbonization. It seems to me that with a temperature
as high as 750° or 800° you would get an appreciable secondary
reaction.
Mr. Parr: I think that in general the point at which naph-
thalene products begin to show themselves is a pretty good line
of demarkation or an indication cf the beginning of secondary
decomposition.
Mr. A. R. Powell: Mr. Chairman, I was much interested
in the results on the sulfur in low-temperature carbonization.
From experiments on laboratory- and plant-scale gas retorts,
I found that the organic sulfur is only partially involved. A
large part is retained in the final coke and the pyrite is decom-
posed. It starts combustion about the same time that the
organic sulfur is evolved. The sulfur in the gas rapidly reaches
a maximum and then falls off, so that in the latter part of com-
bustion, in which we get a gas higher in hydrogen, the sulfur is
very low and there is not a building up of the sulfur later, as
Professor Parr says. I was wondering what the conditions in
low-temperature combustion were that made these results on
sulfur so different from the high-temperature combustion.
Mr. F. W. Sperr, Jr. : Mr. Chairman, I would like to ask Dr.
Parr if he can give some information regarding ammonia. What
amount of ammonia is evolved at the temperature at which he
worked?
Prop. E. P. Schoch: Mr. Chairman, I would like to ask
I 1 Parr to state whether the distillation was carried out by
filling the retort, heating it up and emptying it, as you might
call it, discontinuous; or whether he had a continuous furnace that
was being fed continuously at the top and emptied continuously
at the bottom, since naturally the materials would distil up
in the mass above in one case and not in the other.
Dr. H. L. Olin: Mr. Chairman, I would like to ask Professor
Parr if the low-temperature coke has been examined from the
standpoint .of use in glass furnaces, and his opinion of its value
as a furnace coke.
Mr. Parr: Mr. Chairman, with regard to Mr. Powell's
question as to the behavior of the sulfur, I think we shall have-
to defer our sulfur discussion until some future time. He is
finding out so much about sulfur, and we are also finding out so
many other things, that I am almost persuaded that we do not
know very much about sulfur. I have been taken to task by
somebody — Mr. Sperr, I think — in some of the statements I
have been making recently. I will say only this about sulfur,
and it will partly answer the question about nitrogen. Sulfur
and nitrogen, "and we believe oxygen, practically let go of their
original forms of combination; but just when and how, and
what the conditions are, is a little difficult as yet to understand.
They form in the finished coke new and unusual compound^
which are far removed from what they were in the coal. They
are not chemical compounds in the usual sense, and bear little
relation to anything we know in the way of chemical compounds.
They do not conform to any rule of definite proportion, but come
nearer perhaps to some of the attenuated stages of what for any
better term we may call an adsorbed condition. As an illus
tration, sulfur can be made to unite with a coke which has
absolutely no sulfur in it at all, like sugar carbon, in just about
the amount that we find it in coke. Now, there was no sulfur
in the sugar carbon, but you can make a compound of sulfur and
carbon, stable at 10000, or whatever temperature you choose
Nitrogen behaves in exactly the same way. We can make a
nitrogen carbon at 1000 °, starting with coke that has absolutely
no nitrogen in it. What is this new compound? It isn't an or-
ganic compound ; it isn't an inorganic compound ; and we say
we believe oxygen behaves the same way, and that it is often in
coke as a stable compound up to certain temperatures. I don't
think I care to go into that question, further than to say it is a
field which contains so much yet to be found out that I am
reluctant to venture very far into it. I think Dr. Powell might
perhaps go farther than I would be willing to.
As to Mr. Sperr's question about the ammonia yield, th
actual nitrogen in combination as ammonia is very nearly the
same in amount as is produced from the high-temperature pro-
cess, but it is not due to similar conditions. The high-tempera-
ture process has torn a lot of the ammonia to pieces, and they get
the residue, which is their yield. We do not decompose the
ammonia to the same extent; there is more of it but in other
forms as, for example, the amines in the tars, but the ultimate
nitrogen that we can recover as NH3 is about the same in amount
as from the high-temperature process.
As to furnace methods, this is a discontinuous process, very
much as any coking process is. I doubt very much if our method
would be applicable as a continuous process. We must observe
for the different stages rather exacting conditions, and when the
reactions are completed discharge the batch and begin on a new
lot. We are now attempting to measure the quantity of heat
involved in the exothermic reactions. All we know at this time
is that there is not enough heat generated to do all the work
involved in vaporization of the water, heating up the coal mass,
and supplying the loss due to escaping products and radiation :
hence the intermittent character of the method.
This in general will give you an idea of the procedure. AD
of our work is done with a comparatively small outfit in which
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
17
we can meet these conditions. We work on about 35-lb. samples
of coal.
Just one other word in regard to the inquiry about whether
the material is any good for metallurgical purposes. With Mr.
Sperr in the room, I would hardly attempt to pass judgment
on it in that particular. I asked a blast-furnace man not long
ago to describe what a good blast-furnace coke was. He said,
"After changing our minds so radically within the last few years,
we are not absolutely sure as we once were." Some will say
that it is entirely unsuited for blast furnace use. It does
violence, I think, to nearly everything that would ordinarily
be described by a blast-furnace man as being necessary. I
do not know, however, but that if we could make enough at the
rate of 35 lbs. per run to supply a blast furnace for a week, we
could find out for a certainty.
I want to say in that connection that the initial incentive for
all this work comes out of the great anthracite strike of 1902.
We thought it would be desirable to make smokeless fuel for
domestic purposes out of Illinois coal, and that has been the
main idea all along. We do not know much about metallurgical
coke, although I will say this, that so far as strength, and carrying
the burden, and a lot of those physical conditions are concerned,
it certainly looks very encouraging, but there are other condi-
tions, like high ash, etc., which would enter into the problem.
CARBONIZATION OF CANADIAN LIGNITE'
By Edgar Stansfield
Lignite Utilization Board, Ottawa, Canada
The researches on lignite outlined in this paper
were commenced early in 191 7 by the chemical staff
of the Fuel Testing Division of the Mines Branch,
Department of Mines, Ottawa, and the work is still
in progress. The primary object of the investigation
was to obtain accurate data essential for the scientific
design and control of a plant for the carbonization of
lignite on a commercial scale, rather than to design
such a plant.
In the summer of 1918 the Lignite Utilization Board
of Canada was created by an Order-in-Council of
the Dominion of Canada, supplemented by an agree-
ment as to finances with the provincial governments
of Manitoba and Saskatchewan. The Board was
created to establish an industry for the conversion
of the low-grade lignites of southern Saskatchewan,
and elsewhere, into a high-grade domestic fuel by
means of carbonization and briquetting. The labora-
tory investigations of the Lignite Board have been
carried out at the Fuel Testing Station of the Mines
Branch by members of the staff of the Board working
in cooperation with the members of the Mines Branch
Staff. This latter work has carried to a logical con-
clusion the earlier work of the Mines Branch. The
points essential for the successful carbonization of
lignite, under the economic conditions prevailing in
southern Saskatchewan, were first decided upon,
and then a carbonizer design was evolved which em-
bodied these features. A semicommercial-scale car-
bonizer was erected in Ottawa, and, after many trials
and modifications, successfully operated.
It is worthy of note that the experience and infor-
mation gained in the operation of the carbonizer at
Ottawa have been embodied by the engineer of the
1 Published by permission of Dr. Eugene Haanel, Director, Mines
Branch. Department of Mines. Ottawa. Canada.
board, Mr. R. De L. French, in the design of six
carbonizers for a plant now being erected by the
Board near Bienfait, Sask. This plant is expected
to treat about 200 tons of raw lignite per day.
This paper attempts to trace in outline the progress
of the investigation up to the operation of the car-
bonizer in Ottawa, and to show why this particular
design of carbonizer was adopted. No full report
of any stage of the work has yet been made, but the
methods employed and results obtained in the earlier
stages have been published in some detail.1
The work falls naturally into several stages, but
these are not chronologically distinct. The investi-
gation was commenced with lignite from the Shand
Mine in the Souris, or Estevan area, Sask. Later
other Souris lignites were studied. Now Alberta
lignites, and also peat, are being tested in a similar
manner.
Souris lignite when mined contains from 30 to 35
per cent of inherent moisture, and has a calorific
value of about 4000 cal. per gram. It loses moisture
rapidly when exposed, and the lumps then disinte-
grate. This lignite is employed in the raw state, but
it is a low-grade fuel, unsatisfactory for transporta-
tion or storage. By drying and carbonizing it, a
product is obtained which may have a calorific value
as much as 75 per cent higher than that of the original
coal.
SMALL-SCALE LABORATORY TESTS
In these experiments samples of from 3 to 10 g.
were employed. This allowed very exact control of
the conditions of the experiment, and also allowed a
large number of experiments to be carried out, under
widely varying conditions, within a reasonable time.
It was not possible, however, to study the by-products.
The results were used to cut down unnecessary work
in the larger tests, and were also valuable as checks
on the accuracy of control in all subsequent experi-
ments, and for the comparison of different lignites.
The factors determined included the yield, analysis,
and calorific value of the carbonized residue. The
conditions under which the lignite was carbonized
were varied in order to show the influence on the
results of the final temperature to which the charge
was heated, the rate of heating, the pressure in the
retort, and the atmosphere in the retort.
coal used — -The particular coal chosen for most
of these experiments was from the Shand mine of the
Saskatchewan Coal, Brick, and Power Co., Ltd.
The sample, which consisted of a single lump of coal
shipped by express from the mine in a wooden box,
was crushed and ground to a fine powder in a ball mill.
For convenience of manipulation, and as a preventative
of the rapid change which a powdered coal undergoes
owing to moisture loss and oxidation, this powder
was briquetted in a small hand press. The briquets
were cylindrical, 0.25 in. in diameter, about 0.25 in.
long, and ran about 5 or 6 to the gram. They were
stored in stoppered bottles until required, and from
1 Stansfield and Gilmore, "The Carbonization of Lignite," Trans.
Roy. Soc. Can., [31 11 (1917), 85; [31 12 (1918), 121. See also Mines Branch
Summary Reports for 1918 and 1919.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
Temperature of Carbonization, degrees C
o o o 00000
.'K 1 1 1
^
,'''
*??'"'
\\
\ s
\ \
Carbonization of
western Dominion Lignite **I076
Full curves ~ determined
Dotted curves - calculated results
Analysis of Cool
Raw Dry
Moisture % 313
Ash % 8-0 HO
Volatile Matter % P8 O 4Q-B
fixed Carbon % 3Z7 4T6
Colorific Value c/jg 4190 6OSO
N
\ \
\ \
\ \
\
V
\
N
\
\
N
\
c!N
f'
\ V
\
5500
5000
\
V
y \
\
\
\- -
____.
_o-— "
J^=-
-^J^- — '
\ \
\ \
\ \
\ \
*■*• —
\
S
\
\
\
o\
\
V
.''
,.''
"
\
\
\
'N
\
\
\
\
4000
Yield on coal as charged, per cent
as ao
Yield on dry coal per cent
time to time moisture control determinations were
made upon them. It may be noted that during a
period of 2 mo. the moisture contents fell only 1 per
cent from an original of over 30 per cent.
The gross calorific value of this coal was 4260 cal.
per gram. Its average analysis was as follows:
Per cent
Moisture 31 .8
Ash 5.2
Volatile matter 28.9
Fixed carbon 34.1
apparatus — The apparatus used for most of the
experiments consisted of a cylindrical iron retort
1.5 in. high and 1.5 in. diameter, inside measurement,
having a lid which was held on by a small clamp, the
joint being rendered airtight by means of an asbestos
gasket. A small inlet tube was screwed into the
bottom of the crucible, and an outlet tube into the
lid, the inlet and outlet tubes being so arranged that
the retort could be completely immersed in an oil
or lead bath. For the experiments under pressure a
slightly larger and heavier retort was employed, with a
hexagonal screw cap rendered gastight with an asbestos-
copper gasket. The inlet tube was dispensed with,
and a pressure gage and relief valve connected with
the outlet tube.
method — The coal briquets were weighed out into
a quartz crucible which fitted inside the iron retort.
The heating was done by immersing the retort in a
bath, which for tests up to 3000 C. was of oil, and for
those above that temperature of lead. The lead
was contained in a 4-in. length of 4-in. iron pipe with
a capped end, and was heated in a gas-fired furnace
which gave a very uniform temperature throughout
the bath, and which permitted rapid heating and
easy control. The temperature was followed by two
pyrometers immersed in the lead.
For the regular tests, the retort was plunged into
the bath, previously heated to the desired temperature.
The temperature was kept constant until the evolu-
tion of gas ceased, and the retort was then removed,
cooled, and opened, and the contents weighed and
examined. In other tests, the retort was slowly heated
to about 2500 C. in an oil bath, then transferred to a
just molten lead bath, and the temperature slowly
raised to the desired point. In the vacuum tests,
the pressure in the retort was kept below 25 mm. of
mercury by means of a good water pump. In the
steam tests, a slow current of steam was passed through
the retort. In the pressure tests, the relief valve
was closed at the beginning of the test, but was opened
as required to maintain the pressure in the retort,
due to the escaping gases, at about 120 lbs. per sq.
in. Dry coal was employed for the pressure series.
A striking phenomenon, first observed in connec-
tion with the vacuum series, was later found to take
place with every sample of dried or carbonized lignite.
In every case the residue rapidly gained in weight
after removal from the retort, even when stored in a.
Jan., 1921 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
19
Temperature of Carbonization, degrees C
1
*'*
^s
^'
\ s
7500
MP''
\
gstfi?-
\ °- N
_„•» "
\S\
Carbonization of lignite *I507
from Tofield Coal Cos mine.
7000
\*r-Z.
Tofield, Alta.
\ \
full curves - determined
Dotted curves - calcu/ated resufts
Analysis of Coal
Raw Dry
6500
^s
s
/
\^ \
\
/
\
Moisture % 24 5 -
Nx
tf
?,'
. \
\ \
Ash % 5 6 7 4
Ns
Volatile Matter % 238 335
r,<V
6000
\ \
\ \
Calorific Value ?g 4830 6480
\ \
V
V'
\
5500
•' N
N*
*'
\*,
\ \
\ \
\l—
5000
\
.
k
\\
\
\\
\
\
\
\
\
1
M
4000
100 35
Yield on coal as charged, per cent
\45 40
I0O 95 90 85 80 75 70 65 60
Yield on dry coal, per cent
desiccator over sulfuric acid, its calorific value at
the same time decreasing. This was later shown
to be mainly due to an occlusion of air. All published
results are, with a few stated exceptions, for weights
and calorific values determined immediately after
the experiment.
Figs. 1 and 2 show in graphical form the principal
results obtained in the regular tests on one Saskatch-
ewan and one Alberta lignite.
In every lignite tested the calorific value of the
carbonized residue increases up to a maximum and
then decreases. The temperature for maximum calor-
ific value lies between 550° and 650° C, varying with
the lignite. But the yield of carbonized residue
for maximum calorific value has been found to be
remarkably constant when expressed on the basis
of the dry coal taken. Five out of six samples taken
from different areas in Saskatchewan and Alberta
gave a maximum value with about 67 per cent recovery,
the sixth with about 71 per cent.
LARGE-SCALE LABORATORY TEST
In these experiments the results determined include
the yield and calorific value of the carbonized residue;
the yield, composition, and calorific value of the gas
generated; the yield, calorific value, and economic
value of the tar produced; and the ammonium sulfate
yield available. The conditions under which the
[ignite was carbonized were, in the experiments here
described, varied only to show the influence op. the
results of the final temperature to which the charge
was heated, the rate of heating, and the moisture
conditions of the coal treated. Further experiments
have been commenced which show the effect of the
pressure in the retort and the atmosphere in the retort.
apparatus — The apparatus (Fig. 3) employed in
most of these tests embodies three important features:
(1) Accurate temperature control.
(2) Reduction, as far as possible, of the temperature lag
from the walls to the center of the charge.
(3) Complete removal and easy collection of the tar vapors.
The temperature control is effected by the use of
an electrically heated lead bath, B, with suitable
thermal insulation. The bath rests on a movable
platform which can be raised by the screw C. The
temperature is observed by means of a pyrometer
and regulated by switches and rheostat.
The reduction of lag is effected by the use of a tubu-
lar retort, A. This consists of seven 12-in. lengths of
2-in. boiler tubing, mounted in a cast-iron head. No
part of the charge is thus more than 1 in. from the
walls of the retort, which has a capacity, to the top
of the tubes, of 2300 g. of pea-size lignite with about
35 per cent moisture content. In later work, a cast-
iron retort of cruciform cross-section was employed.
This has a capacity of 3500 g.
collection of tar — A satisfactory method for
collecting the tar was evolved only after many weeks
THE JOURNAL OF INDl STRIAL AND ENGINEERING CHEMISTRY Vol. 13, Xo. 1
of work and many failures. Not only was it hard
to remove the last traces of tar fog, but the condensate
was usually in the form of a watery emulsion, very
difficult to handle.
The method employed was as follows: The hot
gases leaving the retort passed down through the center
tube of a small scrubber, D, made of iron pipe and
containing three interlacing coils of wire, and passed
up again through a surrounding annular space; the
whole scrubber being jacketed with superheated
steam. The heavy tar oils were here condensed in a
practically water-free condition, and dropped into a
weighed glass beaker. The lighter oils, steam, and
gases passed on and down' through the simple tubular
condenser E, where the two former condensed and
collected in a receiver, the oils floating on the water
and showing only a slight tendency to emulsify. The
cool gases leaving the condenser still contained some
tar fog; they were therefore passed down through a
tube scrubber, F, filled with glass beads and a thin
layer of glass wool (shown shaded), through which a
jet of steam from a weighed boiler was also passed.
The bottom half of this scrubber was water cooled.
This scrubber completely removed the tar fog from
the gas. The oil first condensed on the beads acted
as an oil scrubber collecting more of the tar, the steam
prevented the clogging of the scrubber by keeping the
tar hot and fluid, and also, when condensing at the
bottom, carried down with it any vapors still re-
maining. The gases were thus completely cleaned,
and all the liquid products, as well as the ammonia,
from the lignite were collected in the vessels and
could readily be weighed and examined. The tar
thus collected was reasonably free from water and
could be redistilled without excessive bumping or
frothing. The gases leaving the scrubber F passed
through a final cooling tube, G, through a gas meter,
If, ami into a gas holder which is not shown in the
figure.
For temperatures above 7000 C. a smaller apparatus
was employed, with no lead bath. The retort con-
sisted of a simple piece of 3-in. boiler tube, 16 in. long.
It was heated by placing it inside a tube of 3-in. bore
wound around the outside with a coil of nichrome
wire. A charge of 1000 g. was taken for all experi-
ments with this retort. The temperature of the
lignite was observed by means of two pyrometers,
one in the center and one near the wall of the retort.
method — In the regular series of tests, with rapid
heating, the retort was charged, usually with pea-
size lignite containing about 34 per cent moisture,
but in a few experiments with dried lignite, and con-
nected to the purifying train which was then swept
out with gas from a previous run. The lead bath,
heated to a temperature higher than that desired for
the test, in order to allow for the cooling effect of the
retort, was then raised to surround the retort. The
temperatures and pressures at the different parts of
the system and also the meter readings were recorded
at frequent intervals, and the experiments continued
until the evolution of gas had practically ceased.
The gas volumes were corrected for temperature,
pressure, and moisture content, being reduced to
moist gas at 6o° F. and 30 in. of mercury. All other
products were weighed, and all the products were
carefully analyzed. In a number of the experiments
the gas was collected in two separate holders, and the
two portions were analyzed separately. The gas
from the second half of the run is much richer than
that collected in the first holder.
In some tests slow heating was tried, and in others
the retort was evacuated, or was kept under pressure,
or a slow current of steam was passed through.
The results cannot be summarized. The following
are a few of the most important results obtained by
the rapid carbonization of Shand lignite at 555° C.
Weight Balance Sheet (Dry Coal Basis)
Per cent
Water of decomposition 11.7
Gas 17.0
Crude tar 4.1
Carbonized residue 66 . 7
Loss 0.5
Thermal Balance Sheet (Heat Content of
Products as Percentage of Heat in Original
Charge)
Gas 8.3
Tar 6.0
Carbonized residue 78.1
Loss 7.6
Commercial Products (Yields per 2000 Lbs.
of Moist Coal Charged)
Gas, cu. ft 3130
Ammonium sulfate, lbs 10.2
Tar. imp. gal 5.3
Carbonized residue, lbs 910
The coal charged contained 31.8 per cent moisture.
The gas had a gross calorific value of 385 B. t. u. per
cu. ft. and a density of 0.94. The crude tar had
a density of 1.00.
LOW-TEMPERATURE CARBONIZATION BY SHORT EX-
POSURE TO HIGH TEMPERATURES
Figs, i and 2 show that the maximum calorific value
of the residue is obtained by carbonization at a tem-
perature of about 600 ° C. It is clear from the shape
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
*o 2
30%
*s%
?ol
n
1
\v
o\/
4fjS
^alue__-G
V/J'S-
>
-0
if
,' <
uffleat aoO'C
-flayer 1' layer
7. 0 O
7. c A -
X ♦ «
*■ — m —
s of Coal as charged
/n
k"
'^7* x
\^z
AsiU^
""" ^
Ho.slure Z IO
/Isft % 13 5
le Matter % 35 5
Tie Value c/g 5 5 to
■"""*"=r=r7
/ /
/ /
J---/'
\*.
v'-C-t
"2
k
/ /
/ /
/
%■
""A
//
//
of these curves that if lignite is heated in a retort
under the conditions usually met in commercial opera-
tions, with the layers near the wall very distinctly
hotter than those in the center of the charge, no
regulation of the average temperature of the mass
will give a residue with the maximum obtainable
calorific value. The amount which the calorific value
of the residue falls below the optimum will increase
with the thickness of the charge and with the tempera-
ture gradient from the walls to the center.
method — Some preliminary experiments were car-
ried out to test the possibility of obtaining the equiva-
lent of carbonization at, say, 600° C, by short exposure
in a thin layer to a distinctly higher temperature.
Samples of dried Shand lignite, crushed to pass a 10-
mesh screen, were carbonized for a definite number of
minutes in a metal box in a muffle furnace electrically
heated to temperatures of 750° to 8oo° C. The boxes
were 6 in. X 3 in. X 1 in., inside dimensions, of No. 18
gage sheet iron, with loosely fitting lids of the same
metal. When making a test the muffle was brought
up to heat, and the lid of the box was also heated. A
charge either to half or quite fill the box was weighed
out and placed in the cold box. The heated cover
was put on, the box immediately placed on the floor
of the muffle, and the muffle door closed. At the ex-
piration of the desired time, the box with its contents
was removed from the muffle, cooled as rapidly as
possible, and the residue weighed and analyzed.
No great accuracy is claimed for the results, which
are shown graphically in Fig. 4. It is obvious that
the number of experiments should have been con-
siderably increased to render the curves reliable. They
do, however, show that the results of such rapid car-
bonization follow the lines which theory indicates,
and the advantage to be gained by further experiments
was not thought to be commensurate with the work
involved.
Comparison of the optimum results obtained with a
0.5-in. and i-in. layer with those obtained by com-
plete carbonization of the same sample at 5900 C.
and at 600° C, show, as might be expected, that the
yield and composition of the residue is approximately
the same in all cases, but that the calorific values
of 6760 and 6750 cal. per gram obtained with tempera-
ture control, fall to 6690 and 6590, respectively, with
the 0.5-in. and i-in. layers.
BEARING OF RESULTS ON DESIGN OF COMMERCIAL
CARBONIZER
The primary object of the Lignite Utilization Board
is to produce a domestic fuel from Souris lignite. It
is therefore desirable, unless other reasons are found
to outweigh this, to carbonize the lignite in such a
way as to give the residue with the maximum calorific
value. It has been shown that this is accomplished
by complete carbonization at a temperature of about
5750 C, and that the same result can be approxi-
mated by short exposure in a very thin layer to a dis-
tinctly higher temperature. As the object to be
attained is to bring all parts of the mass to the same
optimum temperature, a somewhat thicker layer
continually stirred should give the same result as a
thinner layer at rest. The economic advantage, in
the way of reduction of capital cost of equipment, to
TEE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 1.3, No.'i
be gained by the acceleration of the process by the
use of high temperatures is too obvious to need ampli-
fication.
No increase in the yield of by-products can be at-
tained without a corresponding decrease in the yield
and calorific value of the residue. The gas obtained
at the above temperature is barely sufficient to provide
the heat necessary for the operations of drying and
carbonizing the lignite. The tar yield is also low.
The plant of the Board is situated in southern Saskatch-
ewan, remote from any large center of industry.
Under these conditions it does not appear probable
that, in the beginning of the industry, at least, the
possible profits to be made from the full recovery of
by-products will justify either the capital expenditure
necessary for a by-product recovery plant, or the
depreciation of the carbonized residue by any attempt
to increase the by-products. It is fully recognized,
however, that at a later date with a larger and well-
established industry this policy may require revision.
None of the results obtained give any indication that
the use of vacuum, pressure, steam, or other modified
method of carbonization would have any economic
advantage.
Finally, it has been found that Souris lignite does
not soften or become sticky at any stage of its car-
bonization. This is in marked distinction to the
behavior of bituminous coal, and permits a design of
carbonizer which is simpler and cheaper than can be
employed for the latter material.
DESIGN OF CARBONIZER
The design of carbonizer retort adapted to fulfil
the above conditions is briefly described below. The
actual details of construction are unimportant for the
purpose of this paper. It consists essentially of a
strongly heated surface, or retort floor, inclined at an
angle slightly steeper than the angle of repose of the
crushed lignite. The material to be treated flows
clown the heated surface from a hopper at the top,
passing under a succession of baffle plates, which con-
trol the thickness of the layer. The rate of flow of
the material is controlled entirely by the rate of with-
drawal from the bottom of the retort. This can be
accomplished by any suitable mechanism. The retort
is suitably enclosed at the sides and top, and gas
offtakes are provided in the cover. The thickness
of the layer is controlled by the difference between
the slope of the retort and the angle of repose of the
lignite, by the distance between successive baffles,
and by the clearance between the baffle and the retort
floor. The material is repeatedly stirred by its passage
under the baffles.
The heated surface may be heated from below with
gas. It should be hottest at the bottom of the retort
and progressively cooler towards the top. The
temperature of the lower part of the heated surface
may be as high as the materials of construction will
permit. The regulation of the degree of carboniza-
tion of the lignite is entirely controlled by the time
of its passage through the retort, that is, by the rate
of withdrawal from the bottom.
SEMICOMMERCIAL CARBONIZER
Some experiments have been carried out with a very
small model of the above design. In this model the
working surface varies from 2 in. to 4 in. in width,
is 4 ft. long, inclined at an angle of 45 °, and is electri-
cally heated. The bulk of the experiments, however,
were carried out in a retort approximately 10.5 in.
wide and 10 ft. long. The angle of inclination could
be varied at will, but 45° was found to be satisfactory.
Different materials were tried for the floor of the
retort, but ultimately carborundum slabs were adopted.
Twelve baffles were used in the final arrangement;
these were made of cast-iron and supported from the
floor by means of end plates. The clearance under
the baffles varied from 0.5 to 1 in. The lignite was
crushed to pass 0.25-in. mesh. It was found advisable
to dry it before treatment to a moisture content of
15 per cent or less.
The capacity of the retort varied widely with the
degree of carbonization produced, with the tempera-
ture attained in the gas flue below the retort floor,
and with the moisture in the lignite charge. It may
be rated roughly as equivalent to 200 lbs. of raw lignite
per hour.
The results obtained, with regard to output, ease of
control, and smoothness of operation, were regarded
as sufficient to warrant proceeding with the design
and construction of commercial carbonizers on the
same principle, for a plant capable of treating 200
tons of raw lignite per day.
DISCUSSION
Mr. R. De L. French: That I think is briefly what we have ac-
complished so far. While we do not believe that the work is
at an end, yet it was successful enough in our minds to warrant
us in going ahead with the construction of a plant on a commer-
cial scale. This plant is now under construction. We hope
to have it in operation sometime, and when we do, we hope to
be able to say just what this process will cost in dollars and
cents, and whether or not it is a commercially feasible thing
to carbonize Canadian lignite and to briquet the residue and
sell it as a passing fair substitute for anthracite coal, which a
week ago was selling for $22.60 a ton in the most easterly of the
western cities, and at a higher price further west; I think at about
$27 in Regina last week. Our raw coal will cost us about $1.80
at the mine. As we are in the middle of the field we should
have no difficulty in getting plenty of coal at a low price.
I might say that the lignite with which we are dealing is
probably about as low grade a lignite as we have on this conti-
nent. It has the following analysis:
Raw Lignite
Per cent
Moisture 31.8
Ash 5.2
Volatile matter 28 . 9
Fixed carbon 34. 1
Calories, per gram 4260
You can see it is a very wet lignite and hasn't a particularly
high calorific value. Practically all our work has been carried
out on this lignite because we started with it and because we
wished to compare our results we have endeavored to stick
to it all the way through.
Prof. E. P. Schoch (of the University of Texas, Austin,
Texas, who presented the following resume of "A Process for
the Economic Manufacture of Fuel from Texas Lignite") :
Lignites are characterized by a high water content, the prop-
erty of "slacking" on exposure to air, and a high content of
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
23
carbon dioxide (7 to 8 per cent in Texas lignites). It is this
32 to 40 per cent incombustible volatile matter which causes
briquets made from raw lignite to explode in the fire. Hence
lignite must be retorted to render it fit for briquetting. The
question arises: What is the most economic extent of retorting?
For our experimental study of this question, the lignite used was
obtained in the open market in Austin, but all of it was from the
same mine. The lignite thus obtained was of rather mediocre
quality. To our knowledge better lignite can be obtained even
at this mine and certainly in other localities, but what we used
is representative of much of the lignite now sold in Texas; hence,
the figures presented below may be considered to be safe for all
commercial lignites in Texas, but low for specially good lignites.
In our first set of experiments we retorted lots of 10 lbs. each
in powdered form with constant stirring and fractionated the
gas evolved as the temperature was raised. These experiments
revealed :
(1) The fact that the evolution of carbon dioxide ceases
abruptly at about 525 ° C.
(2) That the per cent by volume of carbon dioxide in the
gas collected up to this temperature is from 23 to 33 per cent.
(3) That the other constituents of the gas evolved up to 525 °
C. have high calorific powers, so that the mixture has a calorific
power of 410 B. t. u.
(4) That all the tar is evolved with this gas.
These results were obtained also with a different kind of a
lignite from a totally different field. The gas fractions obtained
at temperatures higher than 525 ° C. have heating powers of
410 B. t. u. per cu. ft. or less, and the total amount of gas ob-
tainable by retorting a ton of this lignite is not more than 6500
cu. ft. (the lignite from another region gave 6900 cu. ft.),
with an average heating power of the whole gas of 410 B. t. u.
This result is in marked contrast with the 10,000 cu. ft. of 400
B. t. u. reported heretofore.
The coke left after complete retorting has an ash content of
25 to 28 or even 30 per cent and a heating power of 10,000
B. t. u. or below. The relatively poor quality of this coke and
the fact that the gas obtained with it would have to be enriched
to make it fit for "city use" led us to consider the feasibility of
retorting the lignite with a maximum temperature of 525° C.
It was evident that by removing as much as possible of the
large per cent (about 30 per cent) of carbon dioxide from the
gas obtained up to 525° C, its heating power could be raised
substantially, and a simple trial showed that this could be done
readily to such an extent as to make the gas directly fit for
"city use."
To try out this whole procedure on a sufficiently large scale,
we constructed an apparatus which retorted 1100 lbs. of lignite
per 24 hrs. and purified all the gas. The retort was a 6-in. cast-
iron pipe placed vertically and surrounded by a brick furnace
7 ft. high, with gas burners at the bottom. The low temperature
required made it easy to operate in such a manner as not to injure
the iron retort; its life is likely to be great. The amount of
gas obtained was 2250 to 2500 cu. ft. per ton of raw lignite with
a heating power of 525 to 540 B. t. u.; the yield of coke was 900
lbs. of 11,000 B. t. u. (or more!), and the yield of dry tar was
2 per cent. The carbon dioxide was removed down to 2 per
cent by means of potassium and sodium carbonate solution.
Calculation shows that the amount of lignite needed as fuel
for retorting is about 7.5 per cent of the lignite retorted. The
coke comes out of the retort at a temperature just high enough
for briquetting, and not so high as to take fire on exposure to air.
The advantages of this procedure are :
(1) A coke of the highest heating power obtainable.
(2) A gas immediately usable in city mains.
(3) The maximum amount of tar obtainable.
(4) A cheap retort with large capacity, operating under mild
conditions, and yielding the coke at a temperature at which it
can be easily and immediately handled for briquetting.
Prof. Parr: I would like to ask Mr. French if he expects
sufficient binder for his briquet to come from the tars. One
of his numerical factors especially interests us. He says 7 per
cent of heat is lost in the final accounting for the heat. If he
finds it possible to locate, with sufficient accuracy, those per-
centages of heat in the various constituents, and then say pretty
accurately here is 7 per cent of heat unaccounted for, we would
like to know about it. It is one method of getting at the exo-
thermic quantity of heat. Seven per cent of 4000 cal. would
be somewhere within the range where we think the measurement
of quantity of exothermic heat resides. That factor, 7.6, is
exceedingly interesting to our work.
Mr. French : A remark of Prof. Schoch's reminds me I
should mention some things myself. We found exactly the same
things in the beginning of our work that he did. We never
got 10,000 cu. ft. of gas or anything like it. I suggest that some
of those high figures may be due to the method of carbonization,
because I know of one case where a man was actually operating
a carbonizer so designed that they fed moist coal to it. The
moisture that was driven off passed through the hot charge
and what you got was a gas producer on a small scale. This
person may have got 12,000 or 20,000 cu. ft. of gas, but he was
getting it at the expense of his residue. I judge from Prof.
Schoch's remarks that he was primarily after gas. We were
after residue, and it appears that with our own carbonizers we
had just about enough to operate the carbonizers, and not much
more.
Mr. Stansfield ran a series of experiments in the small retorts
under pressure, vacuum, and with a steam atmosphere, but
none of these seemed to show any advantage, and he went back
to practically atmospheric pressure.
In answer to Prof. Parr's question on tars, we took the tar and
distilled it at 325 ° C. On that basis, we got what we called
"available binder," a quantity of pitch representing 2.5 to 3
per cent of the carbonized residue, and that is not sufficient.
It is probably not a quarter of what is required. It takes a
large quantity of binder to make residue briquets, because physi-
cally the residue more nearly resembles charcoal than it does
coke. I imagine it will be similar to some coke which Prof.
Parr has here.
Answering Dr. Porter, the water is the water of constitution.
It is dry coal. It is dried at 105° C, and that is the water left
after drying.
Returning to Prof. Parr, so far as loss of heat is concerned,
I would prefer that Mr. Stansfield should answer that question
himself, because I do not know very much about his calculations,
except that I have a number of them, and I know the loss of
heat always runs around the figures given.
THE COMMERCIAL REALIZATION OF THE LOW-TEM-
PERATURE CARBONIZATION OF COAL
By Harry A. Curtis
International Coal Products Corporation, Irvington, New Jersey
The process herein described was developed for
converting bituminous coal into a uniform, smokeless
fuel resembling anthracite in properties. It was recog-
nized at the outset that the problem was one in which
small-scale tests alone would not yield the necessary
data for plant design, and while much valuable infor-
mation has been secured in small apparatus, the
development of the process has been very largely
through use of commercial-size units. For the past
four and a half years large-scale experimental work
has been carried on in parallel with laboratory tests.
The experimental plant, as finally developed, has a
capacity of about 100 tons of raw coal per day, but
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
a. View of Pla
since it was built only for experimental work, no
attempt has been made to operate all units at capacity.
In the course of experimental tests, the conversion of
coal has, however, frequently reached 40 tons per day,
and recently one of the commercial units was run
continuously for 5.5 mo. without a shutdown. The
experimental plant is fully equipped to handle all the
by-products, and includes a tar distilling unit of 100,000
gal. per month capacity to work up the tar into the
usual crude products.
During the World War construction of a com-
mercial plant was begun, as a government war project.
This plant was eventually completed and half of the
retorts put into operation in June 1920. The usual
minor difficulties of a new plant have been overcome
without trouble and the balance of the retorts are
now being put into operation.
DESCRIPTION OF PROCESS
The essential steps in the process are briefly as
follows:
The raw coal is crushed and subjected to low-tem-
perature distillation in horizontal retorts, the coal
being continually stirred and advanced through the
retort by paddles mounted on two heavy paddle-
shafts running lengthwise through the retort. The
retort is heated externally in a gas-fired furnace, and
the by-products are collected essentially as in coke-
oven practice.
During this low-temperature distillation. 8500 to
9500 F. in the gas phase, the volatile matter in the
coal is reduced from, say, 35 per cent to about 10 per
cent, the resulting semi-coke, being a soft, porous
material considerably different from ordinary coke
in structure. It can be used directly in a water-gas
producer or as a boiler fuel, either hand-fired or with
mechanical stokers. The material is not, however,
in good shape for transportation and marketing away
from the plant. The next step consists in grinding
the semi-coke, mixing it with hard pitch and briquet-
ting. The resulting briquets are somewhat like the
ordinary coal briquets on the market, except that they
burn with but little smoke. The final step consists
in charging these briquets into an inclined retort and
carbonizing them at about 18000 F. for 6 hrs. During
this carbonization the pitch is coked and the volatile
matter in the briquet reduced to about 3 per cent.
There is a shrinkage of approximately 30 per cent in
the size of the briquet and the final product is a hard,
uniform fuel, which burns with an entirely smokeless
flame. Its structure is still markedly different from
that of metallurgical coke, and the fuel burns more
freely than coke.
COALS SUITABLE FOR THE PROCESS
At the experimental plant more than a hundred
coals have been put through the process, and in no
case has it been found impossible to make a satisfac-
tory product. The procedure in briquetting has had
to be varied considerably with different coals, but the
hard, smokeless briquet has finally been produced in
every case.
Since the ash in the coal is accumulated in the prod-
uct, it is desirable, although not imperative, that the
ash in the coal be low. Also, if a high yield of by-
products is desired, a bituminous coal of high volatile
content should be used. The process, however, can
be applied to any coal.
It is, perhaps, of interest to mention that several
lignites have been successfully treated, including those
of Texas, Wyoming, Colorado, Saskatchewan, Japan,
and Brazil.
BY-PRODUCT YIELDS
The yield of by-products in any carbonizing process
will, of course, depend on the kind of coal used. In
Table I is given the average yield of various by-products
from twenty-nine different bituminous coals in which
the volatile matter ran from 32 to 41 percent, averaging
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
2$
37 per cent. These results are from small-scale
testing, charging about 33 lbs. of coal in the retort
and making three to six charges to each test.
In comparing these results with those obtained by
others working on the problem of low-temperature
carbonization, it must be remembered that in the
process under consideration both low- and high-
temperature carbonization are used, and the yields
obtained in the primary or low-temperature carboniza-
tion are augmented by those from the subsequent
high-temperature carbonization of the briquets. It
must also be borne in mind that the pitch, which is
one of the usual by-products, is returned to the process,
and yields, on carbonization, some by-products in
addition to the pitch coke which remains in the briquet.
Table T — Average Results from Twenty-nine
Coals Running over 32 Per cent Volatile
Matter
Average Analysis of Coal (Dry)
Per cent
Volatile 36.9
Fixed carbon 56.0
Ash 7.1
Total 100.0
Sulfur 1.1
B. t. u 13.783
Average Analysis of Finished Briquets
Per cent
Volatile 3.8
Fixed carbon 85.1
Ash 11.1
Total 100.0
Sulfur 0.68
B. t. u 12,874
Yield, per cent 66. 1
Yields of By-Products per Ton Dry Coal
Drv tar, gal 34
Gas, cu. ft 84S7
Ammonium sulfate, lbs 21
Light oil from gas. gal 1 .87
Other tar oils, gal 19.3
Pitch, per cent of tar 43
The by-product yields from the commercial retorts
are a little different from those obtained in the small
apparatus, due in part, at least, to the fact that the
primary distillation in the small apparatus is carried
out in an iron retort, whereas the commercial retort
is lined with carborundum, and in order to get capacity
it is necessary to carry a higher shell temperature in
the retort. This results in a little less primary tar
and a little more primary gas than found in the small-
scale tests.
COMPARISON WITH COKE-OVEN BY-PRODUCT YIELDS
Since coke-oven practice is established and well
known, it is of interest to compare the by-product
yields from this process with those from the ordinary
coke oven. If the two processes be compared for a
high volatile coal, say, 35 per cent, it must be assumed
that the coke oven could handle such a coal, and the
yields given in Table II will, therefore, appear a little
unusual for a coke oven.
A further point must be considered in that while
tar is a normal by-product of the coke oven, it is not,
strictly speaking, a by-product of the other process,
since the tar in the latter case is distilled and the
pitch returned to the process. In order to compare
the two processes, then, it must be assumed that
in each case the tar is distilled, and the pitch in the
Carbocoal process charged against the process. In
Table II this is done, the pitch being taken as 68 per
cent of the coke-oven tar and 50 per cent of the other
tar, these being representative figures in each case.
Table II — Products from One Ton of Dry Coal (35 per cent volatile,
7 per cent ash)
Coke Oven Carbocoal
Coke or Carbocoal 66% ( 1 % volatile) 68% (3% volatile)
Gas, cu. ft 10.000 9,000
Light oil from gas, gal 3
Ammonium sulfate, lbs 20 20
Tar oils, gal 3.8 15
Pitch, gal 8.2 None
While there are a few coals of 35 per cent volatile
which can be coked in an ordinary coke oven, such
as, for example, the Illinois coal recently used in a test
conducted by the Bureau of Standards at St. Paul,
coke-oven practice in general calls for a much lower
volatile coal. Instead of comparing the by-products
from a high volatile coal, as is done above, it is prob-
ably far more significant, economically speaking, to
compare the actual average by-product yields from
coke ovens the country over, with the yields which
the process secures, assuming logically that each pro-
cess will use coals to which it is particularly well
adapted. If we take the coke-oven data as the average
of 7800 by-product coke ovens operating in the United
States in 1917, the following figures obtain:
Coke Oven Carbocoal
Coke or Carbocoal, per cent .. . 71 68
Gas, cu. ft 11,000 (Estimated) 9.000
Light oil, gal 2.4 2
Ammonium sulfate, lbs 19 20
Tar oils, gal 2.3 15
Pitch, gal 4.8 None
In speaking of yields from the process, the particular
coal in question must always be considered. In coke-
Feed Mechak
Primary Retorts
oven practice, the range of coals is rather narrowly
limited and it is, therefore, permissible to refer to
average yields, but in the other process, where the
range of coals is not limited at all, no average or stand-
ard yields can be considered. It is, for example,
quite possible to use a coal yielding 20 gal. of tar oils
per ton, or one yielding 75 per cent of carbonized prod-
uct. In the tables above a coal of 35 per cent volatile
has been taken as one to which the process is particu-
larly well adapted.
INDUSTRIAL PLANT
The industrial plant was put into operation in June
1920. It has a capacity of 500 tons of raw coal per
26
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
day and, besides the main plant, includes equipment
for working up the by-products into the usual crude
products for the market. The coal is mined but a
few miles away, and is a good grade of high volatile
bituminous coal. As the coal comes from the cars
it is dumped into a track hopper and elevated to a
crusher, where it is crushed to pass a three-eighths-
inch screen. It is then delivered to six 80-ton bins
in the primary retort building. There are 24 primary
retorts, arranged in four batteries of six each. Each
retort is about 7 ft. in diameter and 20 ft. long, with
a capacity of a ton an hour. The crushed coal is fed
into the retorts by self-sealing screw conveyors and
is stirred and advanced slowly through the retorts by
a paddle mechanism. The by-products are led off
the discharge end of the retorts and handled as in
coke-oven practice.
.1 IKt
513 k
\J W.4t
ffi*3l
1 - VWTl 4'B-V ■
Top View of Secondary Retorts
The semi-coke which is discharged continuously
from the primary retorts is carried by covered con-
veyors to storage bins in the briquet building. Here
it is ground, fluxed with pitch, and briquetted. There
are two of these roll presses having a combined capacity
of about 24 tons of briquets per hour.
The raw briquets are carried slowly up a long cooling
conveyor to the storage bins at the secondary retorts.
From these bins they are drawn into larry cars and
charged into the secondary retorts. The secondary
retorts are built in two batteries of six and four, ten
retorts in all. Each retort consists of six rectangular
chambers, 21 ft. long and inclined at about 300. with
six charging and three discharging doors per retort,
the capacity of the retort being approximately 15 tons
of raw briquets.
The finished briquets are discharged into steel
quench cars and carried to a quenching and loading
station from which they are finally loaded into railroad
cars.
The by-products from the secondary carbonization
are combined with those from the primary, after a
preliminary cooling. The usual by-product equip-
ment is provided, including a light oil plant, and a
tar-distilling plant.
DISCUSSION
Prof. Parr: Mr. Chairman, I would like to ask Dr. On Us
how nearly the pitch residue from the oil or tar in the process
mi t the requirements of the binder for the briquets.
Dr. Curtis: It is about an even break on most high volatile
coals. The point is not one which bothers us at all. Having
a tar plant as a part of the equipment, we can if necessary bring
in outside tar and distil it at a profit, giving the required addi-
tional pitch. In the case of one plant there is a small shortage
and this is being done. The question of pitch yield depends,
of course, on the coal which is being used in the process.
Mr. Sperr: I should like to ask about the amount of gas
produced. As I understand it, the comparison of the yields
of this process with those obtained in by-product coke-oven
practice was made on the basis of the entire gas production.
That is evidently why the figure of 1 0,000 cu. ft. was given for
coke-oven production. Have you any figures that we could
use to compare the surplus gas produced by this process with
that obtained from the by-product coke oven?
Dr. Curtis: The plant at Clinchfield has not been running
long enough to give an accurate figure, but judging from results
obtained at the Irvington plant it takes about 7000 cu. ft. of
gas per ton of coal to run the process. At Clinchfield we do not
consider gas as one of the salable products of the plant, but in
case a plant were located near a city or industrial center, there
would be a few thousand cubic feet of gas which could be dis-
posed of. The gas yield depends, of course, on the coal used in
the process, and with most high volatile coals is somewhat more
than necessary for the retorts.
BY-PRODUCT COKING
By F. W. Sperr, Jr., and E. H. Bird
The Koppbrs Company Laboratory, Mellon Institl-te, Pittsburgh, Pa.
For nearly two years the production of by-product
coke in America has held the lead over that of bee-
hive coke. By-product coke manufacture is now
firmly established and continually growing, while
beehive coke is certain to decline to a position of minor
importance. Although the bulk of the coke and
gas manufactured in by-product ovens is now con-
sumed by iron and steel plants, there is an increasing
tendency for the by-product coke industry to assume
the position of an independent fuel industry, and its
relations are broadening to such an extent that they
must be considered in the study of almost every phase
of fuel economy.
INCREASING SHORTAGE OF NATURAL FUELS
Among the underlying causes of the many-sided
development of this comparatively new industry,
there is, first of all, the increasing shortage of the
important natural fuels — anthracite, natural gas, and
petroleum. The difficulty of obtaining adequate
supplies of anthracite and the inferior quality of the
material have combined to favor the substitution of
coke. Natural gas finds its most satisfactory supple-
ment in coke-oven gas and has a further accessory
in water gas made from by-product coke. Fuel oil
is being replaced to an increasing extent with tar and
tar oils, while benzene has been successfully intro-
duced as a motor fuel distinctly superior to gasoline,
although on account of the comparatively limited
amount of the former available, there is no question
of competition between the two. The high price and
poor quality of the gas oils now available are having
Jan., 19 2 1 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
27
the effect of discouraging the large-scale manufacture
of carbureted water gas, and, here again, coke-oven
gas appears as the most economical substitute. An im-
portant factor in this connection is the high cost of
labor, which has made the ordinary retort process of
manufacturing coal gas an expensive proposition,
and has forced the artificial gas industry to a recogni-
tion of the advantages of carbonizing coal in relatively
large charges, as is done in the by-product coke oven.
THE BY-PRODUCT OVEN AS A FUEL PRODUCER
With the exception of ammonia and its compounds,
each of the primary products of the modern coke
oven has a technically important fuel value. It is
with the primary products that we are the most con-
cerned. Popular fancy likes to speak of a by-product
coke plant as if it were a factory for dyes and drugs;
but this is, of course, a misconception. In America
it is very seldom that the organization of a by-product
coke plant proceeds farther than the production of
the primary products, and although some of these
products are indispensable to our rapidly growing
American chemical industries, it must be recognized
that, no matter how interesting and important this
sort of utilization may be, it is far outstripped, in
terms of dollars and cents, by the utilization of these
and the other products as fuel.
COMPARISON WITH THE BEEHIVE OVEN
It is of some interest from this standpoint to examine
these fuel values in detail. Such an examination will,
for instance, enable us to appreciate the great economy
of a by-product coke oven as compared with the bee-
hive oven which it is displacing. In coking one ton
of high-grade coal in a beehive oven, the following
fuel must be consumed:
Equivalent
B. t. u. Lbs. Coal
Gas, 11,000 cu. ft 6,160,000 440
Tar, 9 gal 1,401.000 100
Light oil, 4 gal 527.000 38
Coke, 100 lbs 1,300,000 93
Total 9,388,000 671
In coking one ton of the same coal in the by-product
oven, we consume simply: Gas 4300 cu. ft. = 2,408,000
B. t. u., equivalent to 172 lbs. coal. For every pound
of coal coked, the beehive oven consumes 9,388,000
B. t. u., or 33.5 per cent of the heating value of the
coal, while the by-product oven requires only 2,408,000
•B. t. u., or 8.6 per cent.
48,166,719 tons of coal were coked in beehive
ovens in 1918. If this had been coked in by-product
ovens there would have been saved the equivalent
of 11,993,513 tons of coal.
FUEL PROPERTIES OF COKE AND BY-PRODUCTS
Some data regarding the fuel properties of coke,
tar, pitch, and motor spirit (obtained by purifying the
benzenes recovered from coke-oven gas) are given
in Table I, while Table II gives information regarding
coke-oven gas obtained by different operating methods,
as compared with producer gas and water gas made
from by-product coke. The figures in these tables
are given as fairly typical, but there may naturally
be considerable variation, depending upon the kind of
coal used and upon operating conditions.
Table I — Fuel Properties of Coke, Tar, Pitch,
Motor Spirit
Air Flame
Require- *— Temp. ° C—
ment With With Air
Sp. Lbs. per -— B. t. u. per Lb.^ Cu. Ft. Cold Preheated
Gr. Cu. Ft. Gross Net per Lb. Air to 500° C.
Coke 12,900 12,860 132 1875 2085
Tar 1.165 72.7 16,120 15,575 162 1900 2115
Pitch 1.250 78.0 15,660 15,370 '155 1980 2230
Motor spirit. 0.877 54.7 18,060 17,360 176 1915 2165
BY-PRODUCT COKE IN THE IRON AND STEEL INDUSTRY
Although, as has been stated, the use of by-product
coke is rapidly being extended outside of the iron and
steel industry, the bulk of this fuel is still employed
in this industry, largely in the blast furnace and, to
a smaller extent, in the iron foundry. The achieve-
ments in the utilization of by-product coke in the
blast furnace are of the utmost importance from
the standpoint of fuel economy. With modern
methods of manufacture, and with a better under-
standing of the conditions affecting coke quality on
the part of the producer and of the conditions requisite
for efficient utilization on the part of the consumer,
the old prejudice in favor of beehive coke has been
almost entirely wiped out. It has been shown in
regular operation that the consumption of by-product,
coke per ton of pig iron is from ioo to 300 lbs. less
than the consumption of beehive coke, and blast-
furnace managers, as a rule, are now just as favorable
to the use of by-product coke as they were formerly
skeptical.
So remarkable a revolution in both opinion and prac-
tice would have been impossible without the develop-
ment of the modern by-product oven with its flexi-
bility of regulation and its means for exact heat con-
trol at every point. Having such an apparatus, a
proper study could be made of the various factors
affecting the quality of coke by-products, such as the
kind of coal and its preparation, oven dimensions,
and oven operating conditions. Simultaneously, the
effect of variation in coke quality upon blast-furnace
operation had to be determined. It was necessary
to go even farther than this — to break away from
old traditions of blast-furnace practice with beehive
coke and to determine what operating conditions of
the blast furnace would be necessary to give the best
results with by-product coke of a given quality.
It has not always been possible to make this sort of
investigation as a systematic procedure; but our
knowledge of the general subject has been gradually
built up to a point of considerable practical value.
There is still a wide field for further development of
this important subject.
DEVELOPMENT OF OTHER USES
A point which it is especially desired to emphasize
here is that the advances scored in the use of by-
product coke in the blast furnace may be repeated
in other lines of application if similar methods are
pursued. What is especially needed is cooperation
between the producer and consumer of coke, to arrive
at a correct understanding of the requirements for
each particular application. Since we have in the
by-product oven an apparatus of the utmost reliability,
capable of treating a very wide range of coals, the
possibilities of future development in the further
28
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
Table 11 1
■ in, Producer and Water Gap,
Heating Value, Air Requirement,
Flame Temperature
Illumi-
CO3 nants O- CO
Straight coal gas before removing benzenes 2.2 3.5 0.3 6.8
Straight coal gas after removing Denzenes 2.2 2.6 0.3 6.9
Rich coal gas before removing benzenes 2.o 4.3 0.2 6.3
Rich coal gas after removing benzenes 2.6 3.2 0.2 6.4
Lean coal gas before removing benzenes 2.1 2.0 0.3 6.0
Lean coal gas after removing benzenes 2.1 1.0 0.3 6.1
Blue water gas 6.0 ... 1.0 39.0
Coke producer gas (cold) 5.0
Coke producer gas (preheated to 500° C.) 5.0 23 . 0
utilization of by-product coke are very great. One
of the most prominent phases of such development
is in relation to domestic fuel, and the systematic
investigations now being conducted by the U. S.
Bureau of Mines, proving the merit of coke for this
purpose, are typical of what ought to be done in con-
nection with other important applications. There
is no good reason for replacing a single pound of an-
thracite with any solid fuel other than by-product
coke, and there is every reason why the utilization
of by-product coke ought to go much further than the
replacement of anthracite.
Other leading uses of coke, outside of the manu-
facture of iron and steel, are in nonferrous metallurgy,
in the production of water gas, as railroad fuel, and as
fuel for general industrial heating, especially where
the avoidance of smoke is desirable. That quality,
physical or chemical, which is best suited for one
application is not necessarily the best for another.
The iron foundry needs coke of different characteristics
from that required by the blast furnace, and still other
characteristics become essential when we consider the
use of coke in a water-gas machine. These con-
siderations are important in making it possible for a
wide variety of coals, producing cokes of different
quality, to be economically and profitably treated
in the by-product oven.
UTILIZATION OF COKE BREEZE
One of the most interesting developments in fuel
economy resulting from by-product coke manufacture
has been in the utilization of coke breeze — a material
which, not more than a few years ago, was regarded
as nearly useless. This material, containing as much
as 85 per cent fixed carbon (dry basis) and having a
heating value of 11,500 to 12,500 B. t. u. per pound,
was formerly disposed of for filling purposes or else
completely wasted. Of late years, with the develop-
ment of improved stoking machinery, it has been
found possible to burn coke breeze for steam-raising
purposes with a high degree of efficiency, and it is
the general practice for by-product coke plants to
obtain their entire steam requirements from this
fuel. After satisfying plant requirements a surplus
of breeze may still be left for sale, and its utility as
fuel is becoming more and more recognized in the
general market.
TAR AS METALLURGICAL FUEL
The yield of tar obtained in by-product coking
varies with the kind of coal used. It may be as low
as 4, or as high as 12 gal. per ton of coal. With the
majority of coals now being coked in America, the
yield is from 9 to 10 gal. per ton. The use of tar for
fuel, especially in steel manufacture, has rapidly
i pure. Flame Temp. °C.
B t u ment Cu Ft. With With Air
per Cu Ft per Cu Ft. Cold Preheated
H: CHi Nl (Gross) Sp. Gr. I Air to 500° C.
47.3 33.9 6.0 591 0.44 5.08 1865 2095
47.8 34.2 6 0 562 0.42 4.99 1870 2100
46.3 35.0 5.3 630 0.45 S.25 1X70 210(1
46.8 35.4 5.4 605 0.42 5.15 1875 2105
57.0 27.0 5.6 528 (1.38 4.40 1875 2105
57.5 27.3 5.7 497 0.35 4.31 1880 2110
49.0 5.0 305 0.55 2.17 1920 2110
14.0 58.0 128 0.87 0.89 1495 1650
14.0 .... 58.0 128 0.87 0.89 1665 1815
increased during the past few years, and many of the
larger steel companies, operating their own by-product
coke plants, do not sell any of their tar for distillation
purposes, but use it exclusively for fuel. In open-
hearth practice, the consumption of tar per ton of
steel is io per cent less than the consumption of fuel
oil. It is advantageously employed in combination
with producer gas. The resulting flame has a much
better melting efficiency than that of straight producer
gas, and the increase in the capacity of the furnace
is much greater than would be accounted for on the
basis of the heating value of the fuel used. These
considerations are of great moment, in view of the
increasing price of fuel oil, and at a time when the
maximum output per unit of investment is essential.
TAR OILS AN'D PITCH
The various tar distillates have been extensively
used in Europe for fuel purposes; but the demands
for such products in American creosoting and chemical
industries will undoubtedly prevent this sort of utiliza-
tion here for some time to come. There has, however,
been a surplus production of one tar product, namely,
pitch, and its burning warrants some consideration.
It melts readily to a liquid similar to raw tar, and,
with a simple preheating arrangement, could probably
be used in the same way as tar. The employment of
pitch as fuel by direct combustion offers some present
promise, but, in view of the increased demand for it.
particularly in the electrochemical industries, it is a
question whether such application can be counted on
as permanent.
THE BENZENES AS MOTOR FUELS
Although the products from the crude light oils,
recoverable from coke-oven gas, are largely used in
chemical industries, the surplus production of these
materials since the close of the war has required their
sale as motor fuel, supplementing gasoline at an
opportune time. The lower boiling fractions of the
crude benzene (benzene, toluene, and xylene) are puri-
fied and used alone or in mixture with gasoline. This
sort of utilization is very important in Europe, where
there is much less petroleum available than in the
United Spates. Here, even if all our coke were manu-
factured in by-product ovens, the amount of benzene
recoverable would be only about io per cent of the
annual consumption of gasoline. However, the dem-
onstrated superiority of benzene motor fuels over
gasoline gives them considerable local importance in
districts where they are produced.1
1 In a certified dynamometer test by the Automobile Club of America,
90 per cent benzene showed 12.3 per cent less fuel consumption than gasoline.
At the same time the horse power was increased, depending on the speed.
At 2000 r. p. m. this was 19.4 per cent greater than that of gasoline. The
higher ignition point of benzene also eliminates knocking (pre-ignitionl.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
COKE-OVEN GAS
In recent years, an increasing number of by-product
coke plants have been built for the primary purpose
of supplying gas for industrial and domestic con-
sumption. The Koppers oven, using part of its gas
production for its own heating requirements, delivers
a surplus amounting to 60 per cent, or even more, of
the tot»l gas. This surplus is about 6600 cu. ft. per
net ton of coal charged, and, after the recovery of
benzenes, the gas has a heating value of 560 B. t. u.
per cu. ft. The heating value may be increased by
retention of the benzenes, by gas separation, or by
enrichment; but each of these courses of procedure
is, in the long run, uneconomical both to the consumer
and the producer of the gas, and is justifiable only where
arbitrary local standards of high heating values are
enforced. Straight coke-oven gas of 540 to 560 B. t. u.
per cu. ft. constitutes an ideal gaseous fuel for domestic
and industrial heating, and the demand for it is con-
tinually increasing. It is, when manufactured at the
rate of 1,000,000 cu. ft. or more per day, the cheapest
high-grade artificial gas. The carbonization of coal
in bulk, as in coke-oven practice, naturally effects
great economy in fixed charges, maintenance, and
operating labor as compared with the old retort process
for the manufacture of coal gas, while the quality of
the coke produced simultaneously with high-grade
gas is far superior.
Among the principal causes for the rising demand
for coke-oven gas are the increasing recognition of the
utility and convenience of gaseous fuel in general and
the growing shortage of natural gas. The relations
of the centers of production of by-product coke to
districts in which natural gas is largely used are
peculiarly fortunate. Coke-oven gas will be increas-
ingly employed to replenish the depleted supplies of
natural gas in these districts. For example, it has
been shown that the total amount of by-product
■coke-oven gas manufactured in the Cleveland-Pitts-
burgh district, which is the largest natural-gas con-
suming district in the United States, is considerably
more than the annual production of natural gas in the
state of Pennsylvania.
THE COMBINATION OVEN IN RELATION TO GAS SUPPLY
Considerations of this nature have given great
importance to the combination oven, which is the
only type of by-product coke oven that can be economi-
cally heated with either coke-oven gas or producer
gas. If producer gas is used, the entire output of
high-grade gas is rendered available for outside con-
sumption. The combination oven is being generally
adopted by those plants which are built primarily
for gas manufacture. Hitherto, the by-product coke
ovens installed in connection with iron and steel
plants have been designed to use their own gas exclu-
sively, and such ovens cannot be converted into the
combination type without rebuilding. In the future,
however, the price obtainable for coke-oven gas will
make it profitable for iron and steel companies to
build combination ovens whenever it becomes neces-
sary to replace or enlarge existing plants or to build
new plants. Combination ovens have been in con-
tinuous and successful operation in Europe for a
number of years, and one of the several installations
in America has been operating during the past 18
mo., partly on coke-oven gas and partly on pro-
ducer gas, in accordance with the demand for surplus
gas and coke. In considering the possible advantages
offered by the combination oven, it should be pointed
out that it can be heated with producer gas made
either from breeze and other small-sized coke, or from
low-grade coal containing either high ash, high sulfur,
or both. A high percentage of sulfur in the gas is not
detrimental to its use for oven heating. Furthermore,
the combination oven may be heated with blast-
furnace gas, which under certain conditions may be
a profitable procedure.
WATER GAS FROM BY-PRODUCT COKE
The growing importance of gaseous fuels for indus-
trial or domestic heating is such that we must look
beyond the direct production of coke-oven gas proper
and consider other gases that may be made in con-
nection with the operation of a by-product coke plant.
Carbureted water gas is being largely manufactured
from by-product coke to augment the supply of coke-
oven gas; but, as has been mentioned, the unsatis-
factory supply of gas oil has had a discouraging effect
upon the manufacture of this fuel. Blue water gas.
on the other hand, offers considerable promise. It
has a heating value of 300 B. t. u. per cu. ft. and thus
stands midway between coke-oven gas and the low-
grade gases, such as producer gas and blast-furnace
gas. It can be used for a wide variety of heating pur-
poses without the necessity of preheating gas or air,
which is not true of low-grade gases.
PRODUCER GAS AND COMPLETE GASIFICATION
Producer gas manufactured from coke also deserves
some consideration in this connection. Coke producer
gas may be manufactured in connection with the
operation of a by-product coke plant, not only for
heating the ovens, but also for furnishing an additional
supply of gas at relatively low cost to mix with and
augment the supply of coke-oven gas. This, together
with the possibilities offered in the manufacture of
blue water gas, brings up the question of complete
gasification of coal. With a process of complete
gasification which has been urged by many authorities
on fuel economy, the plant would ultimately produce
no solid fuel, but would convert all of the coke into
gas to be mixed with the regular coke-oven gas and
sold. Complete gasification offers more attraction
in rather densely populated industrial districts than
in localities where the gas would have to be distributed
over long distances. There can be no question but
that in the former case it will eventually be under-
taken on a large scale, and it is of interest to know the
amount and quality of the gas that would be produced.
Of course, in each case, allowance must be made for
the requirements of the by-product coke plant with
its necessary auxiliary equipment. If complete gasi-
fication were accomplished with the producer gas
system, the plant would produce 86,100 cu. ft. of mixed
gas per ton of coal having a heating value of 183
30
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
B. t. u. per cu. ft. With the blue water gas system,
there would be produced per ton of coal 33,100 cu. ft.
of mixed gas having a heating value of 380 to 385
B. t. u. per cu. ft. The latter gas would be satis-
factory for all domestic and industrial purposes, while
the former would be of more limited application.
TECHNICAL PROGRESS AND FUEL ECONOMY
It remains to mention very briefly the technical
developments in the by-product coke industry which
have contributed to fuel economy. There is, first
of all, the fundamental heating principle of the oven
with its provisions for economical heat regeneration,
accessibility, and convenient and exact temperature
regulation. This heating principle not only has
effected an improvement in coke quality and saving
of gas over any other oven system previously intro-
duced, but it has also made possible the combination
oven in which the regenerative system is adapted to
the necessary preheating of producer gas as well as air.
The same principle is retained in the new triangular-
flued oven system, and in a new type of gas oven that
is now being introduced.
The use of silica brick in the construction of by-
product coke ovens is now universal in American prac-
tice and has been an important factor in fuel economy.
By its superior heat conductivity, this material has
not only made possible a considerable saving in the
heat requirements of the oven, but has effected a re-
duction in the time required in coking a charge of coal,
and thus has increased the carbonizing capacity per
oven. Its highly refractory quality makes possible
the employment of higher flue temperatures, which
have also contributed to reduction of coking time.
From the standpoint of durability, it is superior to
any other available refractory material. Its use
has an important part in the acknowledged superiority
of American coking practice over European.
Of the number of new developments that are just
at their beginning, there should be especially men-
tioned those that are related to the by-product gas
producer, which is admirably adapted to economical
operation in combination with the by-product coke
plant. The by-product producer is used to a large
extent in Europe; but so far, conditions have not
been favorable to its introduction into America. The
future will, however, see much important progress in
this direction, and it is expected that the same degree
of superiority will be attained as has been achieved in
the introduction and development of the by-product
coke oven.
Work is actively in progress in connection with other
developments and improvements in by-product coking.
One general statement might be made in relation to
these. It has been our experience that improvements
made primarily for the betterment of coke quality
generally have a favorable effect upon the by-products.
In dealing with any given coal supply, it is not at all
necessary to sacrifice coke quality for good by-product
yields, as used to be supposed. This is important
because the profitable disposal of coke is an essential
factor in the success of any enterprise of by-product
coking.
DISCUSSION
Dr. E. W. Smith: Mr. Chairman, Mr. Sperr gave us a very-
low figure, a figure of 8 per cent for fuel oil by-product coking
plants. I should be very glad if he could tell us in connection
with that very low figure what percentage of by-product gas
he used for heating the ovens, and what was the temperature
of the combustion chambers, the volatile matter in his coke,
and the duration of charge. The figures that we are used to
on the other side are figures that are higher than those he has
been fortunate enough to get here. Mr. Sperr will probably
be well acquainted with the fact that the advance that he hopes
to make in this country in by-product producers was made in
Birmingham, England, in 1912, and has worked successfully
ever since. There they have a battery of 66 ovens heated
by means of by-product producer gas, and heated very success-
fully. Those ovens were put in as being the cheapest form of
gas making, because of low labor costs. Since that time, how-
ever, there have been other developments, and that particular
undertaking is installing on wholesale lines the vertical retort,
whtch with slight steaming yields up to about 6000 cu. ft. to the
ton of water gas; gas is made at a cost on a B. t. u. basis (and that is
about the only basis on which we can compare them) much
lower than those obtained from by-product coking, in spite
of the fact that in by-product coking there is a receipt of nearly
one pound per ton more for coke than is obtainable from the
coke from the vertical retorts, so that there are advances being
made in continuous working vertical retort practice of a very
large order, which I think the by-product retort people will
have to watch, if they are going to hold the position that they
have taken in this country.
Coke ovens are being installed here for the purpose of supply-
ing city gas, and the coke used for the production of water gas
and for domestic purposes. In so far as this is true, I am very
strongly of the opinion that gas engineers are not adopting either
the cheapest or the best means of producing city gas. It is an
accepted fact in England that hard coke such as is obtained from
coke ovens or from intermittent verticals does not give anything
like as good results as the special highly porous coke obtained
from continuous working vertical retorts, particularly in water-
gas manufacture.
Domestic coke here is usually hard coke, but when the con-
sumer has been educated into the use of more porous coke, I
am quite satisfied that here, as in England, a market can be
created where this is necessary. The other advantages of in-
stalling continuous working vertical retorts are too well known
to require elaboration and, of course, by-product recovery
is carried out in a similar way to methods employed in coke-
oven practice. I shall be glad if Mr. Sperr can give me those
figures.
Mr. George K. Brown: Mr. Chairman, I would like to ask
one other question: Is it possible to use a vertical continuous
retort similar to the Woodal type as installed by the Porter
Company on a by-product coke? Has it been used, or if it has
not, briefly, why not?
Mr. Sperr: Answering Dr. Smith's question I would say
that the figure for the amount of gas used in coking is based on
the actual operating records of several American plants, such as
the Minnesota By-Product Coke Company at St. Paul, the
Jones & Laughlin Steel Company at Pittsburgh, and the Raiuey-
Wood Coke Company near Philadelphia. I would say in a
well operated plant, not calling for perfection but what you
would reasonably expect in regular operation, you should use
from 38 to 42 per cent of the total gas for coking; the rest you
would recover as surplus. The kind of coal used is an important
factor in the amount of gas required.
The fact that much larger amounts of gas are used for coking
in English practice is due to differences in oven design, to smaller
oven capacities, and to the use of fire clay brick instead of silica
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
31
brick. As a rule, overcoking is somewhat prevalent in English
plants. Of course, in many cases allowance must be made
for the fact that most of the British plants have to use washed
coal, which is charged with a comparatively high percentage of
moisture; but those American plants which also use washed coal
show considerably less gas consumption than the British plants.
Answering the question as to the percentage of volatile matter
in the coal, I would state that this ranges from 31 to 33 per cent
at' the plants mentioned. With the ovens operating at 16 hrs.
coking time, the flue temperatures may be from 2500° to 26000 F.
Now as regards the use of gas producers in by-product coking
practice, we are very glad to give full credit and appreciation
to European technologists for the successful development and
application of the by-product producer. Conditions in Europe
have hitherto been more favorable to the application of by-prod-
uct producers than in this country, but it is certain that the next
few years will witness a great development in this direction here.
With reference to the installation of vertical retorts, adapted
to steaming, Dr. Smith will be interested to know that some of
our newest ovens are also adapted for steaming, and that this
method of increasing the gas production can be employed when
desired. Naturally this is of more interest where the by-product
coke oven is employed primarily as a source of gas than where
coke is the maiu product.
Answering the question of Mr. Brown, regarding the use of
vertical ovens, working on the principle of the continuous vertical
retort, I would say that I do not know of any such ovens that
have been in successful operation. The principle of the con-
tinuous vertical retort is such that it caimot be expected to
produce first-class coke. To attempt to explain the difference
l>etween the functioning of the vertical retort and the functioning
of the coke oven would be rather too long a story for this after-
noon.
Mr. Layng: Are there any ovens in the West using Illinois
coal entirely for coking purposes, and if not what percentage
of Illinois coal may be used in mixtures with Eastern class coals
in the West?
Mr. SpERR: That is a question that always arouses great
interest, particularly here in Chicago. The plant of the Indiana
Coke and Gas Company at Terre Haute, Ind., has used, for long
periods, straight Indiana coal, which is very similar to Illinois
coal. From time to time they have also used varying amounts
of Pocahontas coals in combination with the Indiana coal. These
amounts might range from 8 to 15 per cent. Illinois coal has
also been coked in other by-product plants, either straight or
mixed with different amounts of Eastern coals. I would say
that a large proportion of Illinois coals can be successfully coked
straight in the modern by-product coke oven. The coke has been
found by actual test to be suitable for blast-furnace purposes,
providing the percentage of sulfur is sufficiently low. It is
also adapted for domestic use, for the manufacture of water
gas, and for many other purposes. It is more difficult to make
good foundry coke from Illinois coals, and where the production
of foundry coke is important it is often advantageous to mix
some Eastern coal with the Illinois coal.
The statistics which Dr. Porter includes in his paper for the
year 1917 are, as he explains, not correct in respect to the
present relative proportions of by-product coking and beehive
coking. For nearly two years, beginning, I think, two years ago
this November, the production of by-product coke has been in
excess of the production of beehive coke.
BY-PRODUCT COKE, ANTHRACITE, AND PITTSBURGH
COAL AS FUEL FOR HEATING HOUSES
By Henry Kreisinger
Bureau of Mines, Pittsburgh, Pa.
This paper discusses the comparative value of by-
product coke, anthracite, and Pittsburgh coal, based
on tests made at the fuel laboratory of the Bureau of
Mines, Pittsburgh, Pa. The paper also describes the
methods of firing by-product coke and Pittsburgh coal
that were found to give the best results in actual heat-
ing service.
EXPERIMENTAL
description of fuels — In the tests made at
the Bureau's laboratory, the three fuels were of the
same size, passing over a 0.5-in. screen and through
a 2-in. screen. Their chemical composition is given
in Table I.
Table I — Analyses of Fuels Used in Tests
Proximate Analyses as Received
By-Product Pittsburgh
Constituent Anthracite Coke Coal
Moisture 4.11 0.79 2.23
Volatile matter 6.36 2.80 37.21
Fixed carbon 77.97 79.27 52.10
Ash 11.56 17.14 8.46
I "i\i 100.00 100.00 100.00
Ultimate Analyses of Dry Fuel
Hydrogen 2.58 0.60 5.00
Carbon 82.13 79.24 75.38
Nitrogen 0.87 1.27 1.36
Oxygen 1.32 0.72 7.66
Sulfur 1.04 0.89 1.95
Ash : 12.06 17.28 8.65
Total 100.00 100.00 100.00
Calorific value per lb., as received, B.
t. u 12636 11756 13239
Weights of fuels per cu. f t , lbs 52.5 34 . 5 47.0
The anthracite coal was taken from the Bureau's
stock purchased in 1916. It was a very clean, good-
looking coal, and in fact was considerably lower in
ash than the coal now obtainable on the market.
This fact must be kept in mind when comparing the
results of the tests.
The Pittsburgh coal was sized coal purchased from
a local dealer. It was of average quality as sold in
Pittsburgh.
The by-product coke was a mixture of 60 per cent
of 21-hr. and 40 per cent of 19-hr. by-product coke.
It was made from a mixture of coals coming from nine
different mines. The composition of a composite
sample of these coals is given in Table II.
Table II — Average Composition of Coals Used for By-Product Coke
Constituent Per cent
Moisture 2.77
Volatile matter 34. 17
Fixed carbon 56. 94
Ash 8.89
Sulfur 1.37
Total 100.00
description of tests — The tests were made in two
steam boilers of the size ordinarily used for heating
the average 7 -room house, and were conducted under
conditions conforming to those existing in actual
house heating practice. The tests were started Mon-
day morning and continued through the week until
Friday or Saturday morning. During each 24 hrs.
the fires were run at low rating for a period of 8 hrs.
in a manner similar to that existing over night under
actual heating conditions, and were run the other 16
hrs. to develop a determined percentage of the rating
of the boilers. Three tests were made with each fuel,
one at about 50 per cent, one at 80 to 100 per cent, and
one at 120 to 135 per cent of boiler rating.
On the low rating tests the firings were 8 hrs. apart,
on the medium rating tests about 6 hrs. apart, and on
the high rating tests about 4 hrs. apart. On the
32
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
tests at higher ratings the firings were made closer
together, because not enough fuel could be put in the
furnace to last over longer periods. Between firing
periods the fires were given no attention. Steam
was generated under a 3-lb. gage pressure and dis-
charged into the atmosphere. A large steam separator
was placed in the steam line to take the water out of
the steam. Water was weighed and fed into the
boiler every hour to keep the height of water in the
boiler nearly constant as it would be under actual
heating conditions.
economic results of tests — A summary of the
economic results of the tests is given in Table III.
The third column under each fuel gives the number
of B. t. u. absorbed by the boiler per pound of fuel.
The value per pound of anthracite is high because
the coal contained an unusually low percentage of
ash. Ordinarily the ash in the anthracite runs about
the same as the ash in by-product coke.
Table III — Economic Results of Tests (Averages of Dunning and
Arco Boilers)
, Coke > * Anthracite-
Effi- Ab-
Rating ciency sorbed
52.5 68.9 8105
99.6 70.6 8330
133.0 64.7 7490
Average Efficiency
All Ratings 68.1
Heat Value per Lb.
B. t. u. 11.75
B. t. u.
Effi- Ab-
Rating ciency sorbed
52.9 65.7 8300
89.6 68.40 8640
128.7 66.3 8380
66.8
. — Pittsburgh Coal — -
B. t. u.
Effi- Ab-
Rating ciency sorbed
48.0 55.8 7390
89.6 55.3 7350
108.5 54.4 7200
52.2
The table shows that the efficiency obtained with
the coke was a little better than that obtained with
anthracite coal, and io to 17 per cent better than
that obtained with Pittsburgh coal. The lower effi-
ciency with the anthracite coal is due to the fact
that the coal cracks in the fire and the small pieces of
coal that are cracked off fall through the grate and
increase the losses in the ashes. The low efficiency
obtained with the Pittsburgh coal is due to incomplete
combustion of coal gases and high-flue gas tempera-
tures for a period of i to 2 hrs. after each firing.
If the value of the three fuels is based on the amount
of heat actually absorbed by the boiler per pound
of fuel burned, then the coke is about 15 per cent
better than the Pittsburgh coal, and the anthracite
coal is about 9 per cent better than the coke. However,
as previously stated, the anthracite coal used on the
tests was cleaner than is the coal marketed at present.
With the present market qualities of the two fuels,
the results of the coke and the anthracite coal would
be closer together. Pittsburgh coal is usually low in
ash and high in heat value, so that the comparison
of the coke with the Pittsburgh coal, as shown in the
table, is about right.
No particular trouble was experienced with clinker
on any of the three fuels. Although the coke made
considerable more clinker than either of the coals,
it was light and porous. It formed a circular disk
covering the central part of the grate, and if the fire
was not too hot the whole disk was easily removed
in one piece through the firing door. With a hot
fire the clinker was soft and broke into small pieces
when attempt was made to remove it.
It should be borne in mind that the coke has some
advantages over Pittsburgh coal which cannot be
expressed in dollars and cents. Coke is a clean,
smokeless fuel, requires much less attention when
burned in an ordinary house heating apparatus, and
gives a uniform heat between long firing periods.
ACTUAL HOUSE HEATING TEST
In order to obtain data on the relative value of
coke and Pittsburgh coal under actual heating con-
ditions, the writer used coke at his house during the
months of November and December 191 9, and Pitts-
burgh coal during the months of January, February,
and March 1920. The heated part of the house con-
sisted of 8 large rooms and a bath room. The outside
walls of the house were built of solid concrete with
the wall paper pasted directly on the concrete walls.
On account of this construction the house was rather
difficult to keep warm. The heating plant consisted
of a hot-water boiler rated at 1100 sq. ft. of radiation
surface. The radiating surface of the radiators was
about 600 sq. ft. In two of the upstairs rooms the
heat was turned on about 8 p. m. and off about 7 a. m.
Heat in the other rooms was on all the time. A larger
boiler was installed in order to make it possible to run
the fire with two firings a day; one about 7 a. m. and
the other about 8 p. m. The most important data
for the period between November 1 and March 31 are
given in Table IV.
Table IV — Fuel Used and Weight of Refuse in Heating an 8-Room
House
Wt. of Fuel Wt. of Wt of
f — Burned Lbs. — - Ashes Clinker
Month Day Night Lbs. Lbs. Fuel Used
November 1200 1200 Coke
December 1940 2000 645 245 Coke
January 2890 2000 720 None Pittsburgh coat
February 1981 2162 413 None Pittsburgh coal
March 1570 1545 237 None Pittsburgh coal
During December, when coke was burned, the total
refuse was 890 lbs., of which 645 lbs. were ash pulled
out of the ash pit. The refuse was about 23 per cent
of the fuel fired, and 77 per cent of the refuse was ash.
In January the total refuse amounted to 720 lbs.,
all of which was ash from the ash pit. There was no
clinker. The refuse was 14.7 per cent of the coal fired.
These figures show that the coke had very high per-
centage of ash, which is the principal drawback from
the standpoint of the user. The clinker had to be
removed from the furnace every day or not less often
than every other day. The best time to remove the
clinker was in the morning or in the evening before
firing, and while the fire was not hot. The clinker
could then be removed in one piece, and the removal
was easy. After the clinker was removed the fire
was leveled, and a charge of 60 to 120 lbs. of coke
was put into the furnace. Owing to the greater bulk
of the coke the new charge covered the fire completely,
so that it took an hour or more before all of the new
charge was completely ignited. After the coke once
started to burn a very even rate of heating could be
maintained. The draft needed varied from 0.01 to
0.04 in. of water. The ability to maintain an even
rate of heating depends on the accuracy of draft
regulation. For this reason it is necessary to have
a sensitive draft gage which will easily measure
drafts of 0.01 in. of water. Regulation of draft by the
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
33
position of the damper is unreliable and very unsatis-
factory, and is probably responsible for the many
failures in burning coke. The coke is a clean fuel and
there is no soot deposit on the surfaces of the boiler.
After about 2 mo. of burning coke there was a thin
deposit of fine ash on the surfaces of the boiler varying
from one-thirty-second to one-eighth of an inch in
thickness.
. SECTION THROUGH AB
///J?////////////////////////////;/////////////
With the Pittsburgh coal there was no clinker.
However, to offset this, there was a heavy deposit of
soot on the surfaces of the boiler. If good results are
to be obtained the soot should be swept off of the
boiler surfaces every day or preferably before each
firing. With a proper design of the boiler, the soot
can be swept back into the fire pot, covered with fresh
coal, and burned. It was found that all the soot that
will stick to the surfaces of the boiler will accumulate
in one day. After one day further accumulation is
stopped by the soot burning off. The cubical volume
of one week's accumulation of soot is about the same
as one day's accumulation, but it is somewhat heavier
owing to the fact that a larger percentage of the soot
layer is ash.
The best method of firing Pittsburgh coal was found •
to be as follows: Immediately before firing, the hot
coals were pushed against the rear wall of the fire pot
and the space in the front part of the furnace was
completely filled with fresh coal. In cold weather the
fresh charge completely filled the front part of the
furnace up to the roof of the furnace, even blocking the
door with large lumps. Fig. 1 shows the furnace after
firing.
This method of firing virtually changes the furnace
into a coke oven. The coal in the front part of the
furnace is changed into coke, and the escaping coal
gases pass over the hot coke in the rear part of the
furnace and most of them burn. After 12 hrs., the
coal has been changed into coke; it is then moved
onto the rear part of the furnace and a fresh charge
of coal is put into the front part. The best tool for
moving the coke into the rear part of the furnace was
found to be a spading fork. The prongs of the fork
are inserted between the coke and the lower inside
edge of the firing door frame, and the coke is moved
by a prying motion.
Twelve-hour firing periods are made possible only
with a large furnace with sufficient capacity to hold
enough fuel for 12 hrs.
The writer is of the opinion that heating boilers-
should not be rated on the amount of heating surface
they contain, but on the capacity of the furnace to
hold large firings so that the furnace can be run long
periods without attention. The 12-hr. period is pref-
erable for most houses because the attention the
furnace needs can be supplied by the man, and the
housewife and other members of the family need not
disturb the fires at all.
SOME FACTORS AFFECTING THE SULFUR CONTENT OF
COKE AND GAS IN THE CARBONIZATION OF COAL1
By Alfred R. Powell
Pittsburgh Experiment Station, Bureau of Mines,
Pittsburgh, Pa.
SULFUR IN COAL
It is now known that sulfur exists in coal in three
general forms — pyrite or marcasite, organic sulfur
compounds, the exact nature of which has not yet
been determined, and small quantities of sulfates.
Methods of analysis have been devised for the deter-
mination of these different forms, which have furnished
the basis for investigations of a practical nature on
this most undesirable coal impurity.
Organic sulfur occurs in bituminous coal in quantities
ranging from 0.5 to 2.0 per cent. The quantity
present is very uniform for any given locality and
seam, and it is impossible to remove it from the coal
by any known method. Pyrite comprises practically
all the remainder of the coal sulfur, and the amount
of pyrite present is variable, even in the same mine.
Pyrite may be partially removed from the coal by
washing processes. Sulfates are almost absent in
freshly mined coal, but may increase as the coal stands
in storage.
PRIMARY REACTIONS OF COAL SULFUR DURING CAR-
BONIZATION
A rather detailed study has been made of the changes
these forms of sulfur undergo when subjected to the
coking process. This work has been done in the
laboratory on small quantities of coal in such a manner
that the temperatures could be closely controlled, and
• Published by permission of the Director, U. S. Bureau of Mines.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No.
quantitative study made of the sulfur compounds
in the resulting "products. This would give data on
the primary carbonization reactions, that is, the
reactions without the effects produced by the passage
through the coking mass of volatile matter from
another portion of the charge undergoing another
stage of carbonization.
'Pests on pure pyrite have shown that it is completely
.imposed at 1000° C. The resulting products
are ferrous sulfide and free sulfur, the latter being
I inverted into hydrogen sulfide if hydrogen is present.
A trace of the sulfur remains in the ferrous sulfide
in the form of a solid solution known as pyrrhotite
■ >r magnetic sulfide of iron. The quantity of sulfur
50 remaining, however, is so small that it may be
ed, and the pyritic sulfur may be regarded as
dividing equally between the residue and the volatile
1 1 er of the heated pyrite.
Carbonization tests on a variety of coals have
indicated the five following sulfur reactions:
1 — Complete decomposition of the pyrite to form pyrrhotite
II id hydrogen sulfide. This reaction begins at 300 ° C. is com-
plete at 600° C and reaches its maximum between 400 ° and 500 °
C.
j — -Reduction of sulfates to sulfides. This reaction is complete
» ■ : C.
3 — -Decomposition of one-quarter to one-third of the organic
Milfur to form hydrogen sulfide. This occurs for the most
part below 5000 C.
4 — -Decomposition of a small part of the organic sulfur to
form volatile organic sulfur compounds, most of which find
their way into the tar. - This decomposition occurs at the
lowei temperature of the coking process.
.s — Disappearance of a portion of the pyrrhotite, the- sulfur
apparently entering into combination with the carbon This
reaction seems to be most active at 5000 C. or higher.
The organic sulfur not accounted for by the above
reactions undergoes a decided change in character
between 4000 and 5000, and shows none of the proper-
ties of the original coal sulfur.
These investigations indicate that the total sulfur
of the coal is the most important factor affecting the
sulfur content of the coke, that the relative amounts
of sulfur forms present do not affect it materially, and
that certain other factors, particularly the nature of
the coal, will vary the amount of sulfur in the coke
to a limited extent.
SECONDARY REACTIONS OF SULFUR DURING CARBONI-
ZATION OF CO \I
As hydrogen sulfide travels through the red-hot
coking mass, it is partially converted into carbon
bisulfide. No carbon bisulfide has ever been detected
during the study of the primary reaction.
One of the most important secondary reactions is
that caused by the hydrogen of the gas as it travels
through the red-hot coke. Experiments have shown
that coke practically ceases giving off hydrogen sulfide
after the temperature has passed 600 ° C. However,
if hydrogen or gas containing hydrogen is passed
through coke above 6oo° C, a further and very de-
cided evolution of hydrogen sulfide is obtained.
Two important changes are caused by the passage
of hydrogen through the coking mass:
(1) FeS2 is caused to decompose at a lower tempera-
ture, the decomposition being practically complete
at 500°, whereas in the primary reactions it is only
partially decomposed at this temperature. The
net result of this is the speeding up of a reaction which
would be complete at the end of the coking process
without the hydrogen effect.
(2) The decomposition of the organic sulfur or
"carbon-sulfur" combination of the coke to form
hydrogen sulfide is enormously increased at tempera-
tures above 5000. This means that where the hydro-
gen from the distillation comes in contact with the
red-hot coke, this coke will contain less sulfur than
the primary reactions alone would indicate.
Experiments have been performed to determine
the equilibrium between the sulfur in the gas and the
sulfur in the coke. Hydrogen over a coke containing
r.2 per cent sulfur was found to reach saturation when
it contained about 0.25 lb. of sulfur per M. cu. ft.,
when the coke was at a temperature of 900° C. This
indicates that large quantities of hydrogen would be
required to remove an appreciable amount of sulfur
from coke. The reaction appears to go to equilibrium
very quickly, however. The essential conditions for
the transfer of the coke sulfur and the gas, therefore,
would consist in the passage of hydrogen through the
coke mass at a rapid rate.
These laboratory data on the effect of hydrogen on
the sulfur of the coke were well confirmed by large-
scale practice. Coke obtained in the laboratory, where
the by-products were swept away as fast as formed,
contained a larger percentage of sulfur than coke
made from the same coal at the same temperature
in by-product ovens, where the hydrogen-containing
gases had relatively long contact with the hot coke.
Experiments have shown that by-product coke-
oven gas. purified from sulfur, when passed back
through the oven, causes quite a marked decrease
in the sulfur content of the coke. The unpurified
gas, however, contained sulfur in excess of the satura-
tion point, and actually increased the sulfur in the
coke to some extent. This shows that the passage
of sulfur from coke into the gas may be reversible
under these conditions. These facts bear out the
laboratory data on the effect of hydrogen on the coke,
as well as confirming the fact that an equilibrium
point exists beyond which no sulfur is transferred
from the coke into the gas.
DESULFURIZATION OF COKE
The very interesting fact that hydrogen has such
a desulfurizing effect on coke brings up the question
as to a possible practical application. With this idea
in mind, work is being continued on a study of the
equilibrium relations between the sulfur in the coke
and the sulfur in the gas at different temperatures
and with different percentages of hydrogen. Large-
scale tests are also being conducted to determine how
much desulfurization is possible, as well as to get cost
data on any possible process.
With the supply of low-sulfur coals getting lower,
it has been stated that a reduction of the sulfur
Jan., 1021
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
35
content of the coke to the extent of 25 per cent will
increase the value of the coke $1.00 per ton. In the
laboratory, this figure has been greatly exceeded,
while in actual practice it has been approached. The
value of such a process, if developed to a commercial
scale, would be worth millions of dollars to the metal-
lurgical industries of the country.
DISCUSSION
Dr. Smith: It occurs to me that the Fourth Report of the
Gas Investigation of the Institution of Gas Engineers and the
Leeds University might prove of great interest to Dr. Powell.
They have carried out near Glasgow a long series of tests on
Scotch coals on the vertical retorts, both with steam and without,
and they have obtained figures of a complete balance sheet for
sulfur in all by-products of the coal, for nitrogen, carbon, and
heating value, and I think it will be of interest to Prof. Parr
to know that through all of the figures they have found that
there is an unaccounted-for heat loss of between 3 and 4 per cent,
practically constant, which may be accounted for, as suggested
by Prof. Parr, as having something to do with the exothermic
reaction of coal; but if Prof. Parr cares to see a copy of this report
I shall be very pleased to let him have it. I am sure the figures
there are, from an English point of view, classical.
Prof. Parr: Mr. Chairman, these figures are exceedingly
interesting. Now that I see them, it seems to me this work
is much more in accord with our conclusions than I had thought.
I would suggest as an explanation where there seems to be a
difference, that in Mr. Powell's apparatus there is almost lacking
that condition of purification in the delivered sulfur, for instance,
that we have in the coking chamber with a lot of coal. For
instance, as an illustration, if we take a coke in which there
is an absence of sulfur and pass hydrogen sulfide over it, it
purifies the gas and contaminates the coke; but in an apparatus
of this sort, if you get your products out of the way without
any of that reaction, you will get results which seem to be a
little different from ours. As a matter of fact, they are very
concordant, because, although you notice the decomposition
of the pyrite at a point where we say our gas is pretty nearly
free from sulfur, that is simply because of the powerful action of
that temperature, most active at about 500°, which contami-
nates the coke and purifies the gas, and it is quite in accord with
this chart.
The interesting thing in Mr. Powell's experience, and ours
too, is that this adsorption (for want of a better term) in the
coke reverses at higher temperatures so that its vapor pressure
is such that it can be given off slowly. Assuming that hydrogen
would do the same thing, it would take the place of sulfur, and
we can remove practically all the sulfur content in these arti-
ficially made sulfides of carbon, if that is a good name for them;
also your vapor will do the same thing, and I think that is an
exceedingly interesting phase of the work. I hope Mr. Powell
will follow it up, because I do believe that there is a possibility
of doing these things successively, first purifying the gas and
then purifying the coal. Now go on and collect the sulfur and
we shall have the circuit complete.
THE DISTRIBUTION OF THE FORMS OF SULFUR IN
THE COAL BED1
By H. F. Yancey and Thomas Fraser
Mining Experiment Station, U. S. Bureau ok Mines, Urbana, Illinois
The purpose of the work described in this paper was
to study the distribution of pyritic and organic sulfur
'in coal as it occurs in various sections, layers, or benches
1 Published with the permission of the Director, U. S. Bureau of Mines.
Abstract of a bulletin to be published by the University of Illinois, Engi-
neering Experiment Station; by permission of the Director.
of the coal seam. Sulfate sulfur was entirely disre-
garded because it was found to be very low in freshly
mined coal. It is well known that the variation of
total sulfur between sections or benches of the same
bed at a given place, in any except low sulfur coals, may
be quite marked. This is due principally to the heteroge-
neous or "spotted" distribution of iron pyrite. More
or less of the pyrite, depending on its physical form,
can be removed by coal-washing methods. This
brings up the question of the variations of organic
sulfur content.
Until recently no very satisfactory methods for
the determination of pyritic and organic sulfur in coal
have been available. Parr and Powell1 have given
very satisfactory methods for these determinations.
Wibaut and Stoffel,2 working in the Municipal Gas
Laboratory at Amsterdam, have also developed
methods recently, but those of Parr and Powell have
been used for this study.
While little or no information on this subject is
available in the literature, some previous work3 led
to the tentative conclusion that the organic sulfur
content of a given coal varies but little, and that at
least it is much more uniform than the pyritic and
total sulfur values. One of the objects of the present
work was to determine whether this is the actual
condition, or whether organic sulfur is segregated as
is pyritic sulfur. In case segregations or concentra-
tions of organic sulfur were found to exist, it would
be desirable to associate such occurrences with other
impurities or specifically recognizable conditions. If
organic sulfur segregated, it might then be possible
to remove some of it in the way that pyrite is removed.
METHOD OF SAMPLING AND ANALYSIS
It seemed that the only way to study the subject
was to take channel samples in the mine at the working
faces. Samples have been taken in three mines.
Seventy sectional bench samples were taken at twelve
working faces in the Middlefork mine of the U. S.
Fuel Co., near Benton, Illinois (No. 6 seam). Forty-
eight samples were taken in two mines in western
Kentucky operating in the Kentucky No. 9 and No. 12
seams.
At each place in the mine selected for sampling, the
coal face was marked off before cutting the samples
into from four to eight horizontal benches, and each
bench was sampled separately, according to the
Bureau of Mines method for sampling coal in the
mine.4
Total sulfur was determined by the method of
Eschka. Pyritic sulfur determinations were made
according to the method of Powell with Parr.6 The
values given for organic sulfur represent the difference
between total and pyritic sulfur. Only a few samples
were examined for sulfate sulfur. The highest value
obtained was 0.04 per cent, and this was on a sample
1 University of Illinois, Engineering Experiment Station, Bulletin 111
(1919), 44.
2 Rec. trav. chim., 38 (1919), 132.
8 Thomas Fraser and H. F. Yancey, Am. Inst. Mining Eng., Bulletin
153 (1919), 1817.
« J. A. Holmes, Bureau of Mines, Technical Paper 1.
* J.oc. cit.
36
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 1.3, No. 1
containing 6 per cent total sulfur which had been
mined 3 mo.
DISTRIBUTION OF FORMS OF SULFUR
The distribution of the forms of sulfur at a few
locations in one of the beds examined is shown graphi-
cally in Figs. 1 to 6. Distance from the top or roof
of the bed is represented on the ordinate axis and
the per cent of total, pyritic, and organic sulfur as
abscissas. The vertical lines showing per cent of
sulfur represent the average values for the forms of
sulfur occurring in a section of the length of the line.
The breaks are due to variations in the averages for
adjacent sections, and do not indicate that the sulfur
content changes abruptly at the point of the break.
The sulfur percentages plotted in the graphs represent
values for moisture-free coal.
per ctMT sulpur eta CENT SULFUR
II
1
TOP
!
]
S r
1
i-n
■;
—,
Bottom
Fi&l. I0°N.|SWN
PER CENT SULFUR
2 3-
I
1
TOP
-
1
i
:
SOTTOH ]
Fig. 2. 7HN,ICWN
PER CENT SULFUR
2
3 A
T
!
rw
i
1
hi
DM
Bor
nop
—
1
1
. 1
- -
~|
"i
- I
1
I -
1
"-
1
FlG.3. 4^N, Ist WN
PER C6NT SULFUR
j
1, r
J
j
1
tH l
! 1
-.,
t-f
I-
-'
BO
Fig. 4. /'Iain /North
Distribution 6f
Forms of Sulfur
in the Coal Bed'
Legend ,_'
I Pyritic Sulfur S
i Organic Sulfur
'; Total Sulfur
Fig. 5. 3"-»N, l5-TEN
p
a c
SULFUR
3
TOP
.
J/.
-;
;
i-1
1 1
BOTTOM
Fig. 6. I0T-=N, I'-'ES
In the Middlefork mine, No. 6 bed, represented by
these graphs, the total and pyritic sulfur at most of
the places sampled was higher in the top coal and
bottom coal than in the intervening part of the seam.
This was also true of the No. 12 seam examined in
Kentucky. In the No. 9 seam in Kentucky the bottom
coal was highest in total and pyritic sulfur, and the
top coal was lowest. The organic sulfur content,
on the other hand, shows no large variation between
different benches of the bed at any place sampled,
although, as shown in the graphs, it does not run
absolutely uniform. A closer approach to uniformity
for the values for organic sulfur, between the benches
at a given location, is not obtained by calculating
organic sulfur content on a moisture-, ash-, and pyritic-
sulfur-free basis. The general tendency at the places
shown in the figures which represent the north side
of the mine at Benton, Illinois, is for the organic sulfur
to decrease with increasing pyritic sulfur content. It will
be observed that on the individual graphs, where the
pyritic sulfur in any particular bench is higher than
in the bench adjacent above or below, the organic
sulfur is in most cases lower* This can hardly be
interpreted as supporting the idea that organic sulfur
contributes to the formation of pyritic sulfur, however,
for this tendency is not nearly so evident in the other
half of this mine or in the other two beds examined.
In order to secure additional data on the possible
relation of organic sulfur to pyritic sulfur, a number
of special samples were taken of coal immediately
surrounding or interbedded with bands or cat faces
of pyrite. These samples were found to be about
average or below the average in organic sulfur content.
There is no evidence of a concentration of organic
sulfur in the coal immediately adjacent to pyrite
deposits.
ORGANIC SULFUR IN VARIOUS COALS
The relatively high proportion of sulfur in the
organic form occurring in many coals has not been
generally recognized. It has often been considered
as constituting a negligible percentage of the total
amount of sulfur present. In estimating the wash-
ability of a coal the organic sulfur content is an im-
portant consideration. In thirteen out of the thirty-
four bench samples represented in the figures, the
organic sulfur exceeds the pyritic sulfur content.
This was true of twenty-three out of thirty samples
taken in the No. 12 bed of western Kentucky. Table I
shows the proportion of organic sulfur in samples of
a number of well-known coals.
Table I — Pyritic and Organic Sulfur in Various Coals
(Values in per cent on moisture-free basis)
Organic Sul-
fur as Per
cent of
Total Pyritic Organic Total
Location of Mine Coal Bed Sulfur Sulfur Sulfur Sulfur
Mahaffey. Pa.' C&D .148 .'.77 0.71 20.4
White Co.. Tenn... . Sewanee 4.87 3.59 1.17 24.0
Pike Co., Ky Freeburn 0.46 0.13 0.33 72.0
Herrin, 111..' No. 6 1.83 1.04 0.79 43.2
Greene Co., Ind No. 4 1.66 0.89 0.77 46.4
Benton. Ill No. 6 3.29 1.99 1.30 39.5
Western Kentucky . . No. 12 1.48 0.70 0.78 52.6
Western Kentucky. . No. 9 3.46 1.65 1.81 52.5
McDowel Co., W.Va« Pocahontas 0.55 0.08 0.46 83.7
No. 3
Letcher Co . Ky.'... Elkhorn 0.68 0.13 0.51 75.0
I H F. Yancey and Thomas Fraser, Coal Industry, 3 (1919), 36.
' A. R. Powell, This Journal, 12 (1920), 889.
FORMS OF SULFUR IN RAW AND WASHED COAL
It is evident that if organic sulfur segregated with
or was concentrated around pieces of pyrite, bone
coal, or shale of higher specific gravity, it would be
removed with these impurities as refuse in the washing
operation. On the contrary, removal of the non-coal
impurities, inorganic in nature, should result in a
slight increase in the organic sulfur content of the.
washed coal, depending upon the amount of inorganic
impurities removed. In order to obtain data on this
question, seven samples of run-of-mine coal were
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
obtained at the Middlefork mine, in addition to the
face samples collected in the mine. A large coal-
washing plant is maintained at the mine for washing
the entire tonnage. Each sample represents a day's
production for the mine, which varies between 2400
and 2800 tons. A sample of washed coal representing
one day's operation of the washery, on an average
day, was also obtained. The sulfur forms in these
samples are shown in Table II.
Table II — Forms of Sulfur in Raw and Washed Coals
(Values in per cent on moisture-free basis)
Total Pyritic Organic
Sample No. Sulfur Sulfur Sulfur
72 Raw coal 3.68 2.42 1.26
73 Raw coal 3.20 1.90 1.30
74 Raw coal 3.22 1.99 1.23
75 Raw coal 3.59 2.07 1.52
76 Raw coal 3.33 1.93 1.40
77 Raw coal 3.27 2.08 1.19
78 Raw coal 2.77 1.55 1.22
Average of raw coal 3.29 1.99 1.30
Average of face samples for
mine 3.30 1.92 1.38
Washed coal 2.25 0.92 1.33
Refuse 13.45
The values for organic sulfur in the average for the
run-of-mine raw coal, and in the washed coal are nearly
identical, namely, 1.30 per cent for the raw coal, and
1.33 per cent for the washed coal. Though the washed
coal sample does not necessarily represent the product
obtained by washing the identical coal of the run-of-
mine samples, it must be taken as further evidence
to show that organic sulfur is not segregated with or
concentrated around the high specific gravity pieces
of pyrite, nor is organic sulfur removable by gravita-
tional methods. The average values for the sulfur
forms in the run-of-mine raw coal are in close agreement
with the average for the sectional face samples col-
lected in the mine.
CONCLUSIONS
1 — Extreme irregularity of distribution is charac-
teristic of the pyritic sulfur of coal. This offers a
possibility of securing a low sulfur product by separate
mining of parts of the seam.
2 — In comparison with the large variations of pyritic
sulfur in the vertical span of the bed, the organic
sulfur is quite uniform.
3 — There is little evidence of a definite relationship
in the occurrence of organic and of pyritic sulfur.
High pyritic sulfur in a bench or section of the bed is
not indicative of high organic sulfur content.
4 — The proportion of the sulfur that is in organic
combination in various raw coals varies within wide
limits. High sulfur coals are ordinarily higher both
in organic and pyritic sulfur than low sulfur coals,
though organic sulfur makes up a greater percentage
of the total sulfur in the case of low sulfur coals
(Table I).
5 — The organic sulfur content of some coals is
sufficiently high to limit seriously the extent to which
these coals can be cleaned of sulfur by washing.
ACKNOWLEDGMENT
This investigation was carried out under the general
direction of Mr. E. A. Holbrook, Assistant Director,
and Mr. Geo. S. Rice, Chief Mining Engineer, U. S.
Bureau of Mines. To them and to Professors S. W.
Parr and H. H. Stoek, of the University of Illinois,
grateful acknowledgment is made. Mr. C. A. Meissner,
Chairman of the Coke Committee, U. S. Steel Corpo-
ration, and Mr. Thomas Moses, General Superin-
tendent, U. S. Fuel Co., have followed the progress of
the work with cordial cooperation.
COLLOIDAL FUELS, THEIR PREPARATION AND
PROPERTIES
By S. E. Sheppard
Research Laboratory, Eastman Kodak Co., Rochester, N. Y.
"Colloidal fuels" is the name given to a distinct
class of liquid to semiliquid blended fuels. They
were developed in this country during and subsequent
to the last two years of the Great War. In physical
consistency they range from liquids with a viscosity
at normal temperatures of some 30° Engler to very
plastic pastes, and weak jellies, these latter becoming,
however, relatively mobile and fluid when heated.
They are composites, in which either finely divided
carbonaceous solids or semisolids, or both, are so
suspended in and blended with liquid hydrocarbons
as to form relatively stable and atomizable fuels. They
have been developed primarily for burning with the
regular types of atomizing burners using ordinary
fuel oils, but have also possibilities for use in internal
combustion engines of the Diesel and semi-Diesel type.
WHY COLLOIDAL?
It may be said that there is nothing in this outline,
description to warrant the term "colloid." The term,
however, has a considerable elasticity. I do not
propose to add to the excess of definitions of colloids;
but will note two recent ones. According to Dr.
Wiley, colloid chemistry is the chemistry of "matter
without form and void," and is mentioned in the
first chapter of Genesis. This gives it a respectable
antiquity, and a latitude sufficient to embrace anything.
As against this universal scope, Professor Bancroft
tells us "it is the chemistry of finely divided masses,
in other words, of bubbles, drops, grains, filaments,
and films," and this more specific dictum is certainly
applicable to the systems under discussion. However,
without striving for a dictionary precision, it may be
said that the term is conveniently employed to describe
the product, both owing to certain of the fuels' impor-
tant colloidal characteristics, and because the process
of preparation may be justly termed "colloidalizing,"
in view of its essential dependence upon colloid chemical
processes and conceptions.
HISTORICAL
Before entering into details of the application of
colloid chemistry to the fuel problem, let me say a
few words on the history of the present class of ma-
terials. Theidea of burning a suspension of carbona-
ceous matter in mineral oils appears to be nearly
as old as the use of fuel oil, but no attempt appears
to have been made to investigate systematically its
possibilities.
The developments now described date from the
summer of 191 7. At that time a fellow-worker in
this laboratory, Mr. J. G. Capstaff, asked the author
38
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
as to the possibility of the use of powdered coal in
conjunction with oil to supplement the latter for oil-
burning ships. Actually, an adequate supply of fuel
oil was no less vital to the Allies than gasoline and
lubricants. The German submarine campaign was
threatening all of these. Having much faith in the
possibilities of colloid chemistry, the author prepared
some composites. They contained up to 30 per cent
of pulverized coal incorporated by a paint mill with
an ancient specimen of oil from a laboratory oil bath —
plus one or two things thrown in for luck. These
composites appeared promising as regards stability,
and we succeeded in burning them satisfactorily in an
air-pressure oil-fired furnace. Through Dr. Mees
these results were referred to Mr. Lindon W. Bates,
Engineering Chairman of the Submarine Defense
Association, who was already devoting his attention
to this very problem. At his instance and with
Mr. Eastman's sanction, the possibility of colloidally
combining pulverized coal and fuel oil was taken up
by the research laboratory, in close and constant
cooperation with the Submarine Defense Association,
under Mr. Bates' coordinating leadership; but for this,
and without his catholic knowledge and experience
of fuels and fuel problems, our initial experiment
would probably have remained a laboratory incident.
By "colloidally combining" is to be understood "stably
dispersing pulverized coal in fuel oil," that is, forming
a uniform composite,-the stability of which at ordinary
temperatures should be reckoned in months, while
amply sufficient at higher temperatures to permit
atomization by fuel oil burners. As stated, Mr. Bates
had already been actively considering the possibility
of supplementing oil for marine purposes by pulverized
coal, or oil and coal combined. The Association had
had assigned by Admiral Benson, Chief of Naval
Operation, the U. S. S. Gem, which was operated under
Mr. Bates' direction for research work during the war.
She was fitted with the highest class Normand destroyer
boilers. Whatever the ultimate rating of colloidal
fuels in commercial practice, the technical objective
was effected when, from April to July 1918, this
craft was successfully operated on a colloidal fuel,
containing 30 per cent pulverized coal, as efficiently
as with regular fuel oil. I shall return to these trials
in dealing with the properties of colloidal fuels. It
must be remembered that where a new paint or varnish
requires pounds and gallons for practical trial, a fuel
requires tons and tank loads. Much of the technology
of preparation and control had to be remodified as
the amount prepared increased to this scale, and in this
connection I take pleasure in referring to the constant
and invaluable help of my associate and assistant
chemist, Mr. L. W. Eberlin. First let us consider
briefly some chemical and technical aspects of their
preparation.
SOME PARADOXES OF COLLOID CHEMISTRY
In many ways the science of colloids is a science of
paradoxes. So much is evident in its development.
As is well known, the term colloid was first applied
by Graham to a group of substances, such as gelatin,
starch, silicic acid, or white of egg. He contrasted these
with crystalloids such as sugar, salt, etc., because of
their low or negligible diffusibility, difficulty in assum-
ing definite crystalline form, and relative chemical
inertness.
Graham grouped these properties under the con-
ception that colloids had inergia, that is, an inertia
of energy which made their state at any moment
dependent upon their previous history; whereas the
state of a crystalloid at any moment can be defined
without reference to its history, but is completely
defined by quantities independent of duration pre-
vious to that moment. He considered that they
formed a dynamic state of matter as compared with
the static state of crystalloids. And he believed that
this depended ultimately upon a difference in the mole-
cules of colloids, a greater content of idiochemical
affinity. Paradox shows itself now. The develop-
ment of colloid science in the last twenty years has
been toward quite opposite conclusions, on the whole.
It has been in the direction of regarding colloids as
physically rather than chemically specific. Briefly,
it is argued that any substance in the solid or liquid
state can be brought to the colloid condition if it be
mechanically subdivided so that its particles or drop-
lets are approximately between in and ifi/j. in diameter,
that is, less than 0.00001 cm., but greater than
0.0000001 cm., and kept so in suspension in an indiffer-
ent medium. In terms of this conception, colloids
form a particular intermediate region of dispersed
systems or dispersoids, expressed in the table:
Coarse Dispersoids
Diameters greater than
0.1 p, do not pass fil-
ter paper, can be re-
solved with micro-
scope (up to 2000)
Dispersoids
Colloids
Increasing Dispersity
1 fi ' o 1 fin, pass through
filter paper, not micro-
scopically resolved, do
not dialyze or diffuse
>
Molecular Dispersoids
Diameters smaller than
1mm, pass through
filter paper, not mi-
croscopically re-
solved, diffusible
and dialyzable
True solutions
It is admitted explicitly that the boundaries are
not sharply defined, but that we have a gradation.
It will be seen that this relatively clear-cut con-
ception marks a great change. Colloids and crystal-
loids are not antithetic, but connected by continuous
transitions. The crystalloid condition, involving di-
rected symmetry relations in space, is an internal
molecular condition; the colloid state is an external
one, depending upon the subdivision of multimolecular
masses, and possible to all chemical substances. The
properties of colloids, on this view, depend chiefly
upon the large accession of surface energy, parallel
with dispersity. Dispersity is defined most generally
„ total surface „,, , , ,
as ratio of , . Thus, a sphere has a lower
total volume
specific dispersity than a cube of the same volume,
because its surface is smaller in proportion to its volume.
A large number of properties of colloids can be
explained very reasonably on the view that spontaneous
changes in dispersoids will be in the direction of re-
ducing the dispersity, thus diminishing the free sur-
face energy, and by the conception of adsorption,
i. e., of surface concentration of (molecularly) dis-
solved substances on dispersed material. So far so
Jan., 1921 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
good. But paradox again asserts itself. Considera-
tions of this type are found to be most satisfactory
when applied to so-called suspensoids, i. e., colloid
or pseudo-colloid systems in which no intimate rela-
tion exists between the dispersion medium (solvent)
and the dispersed substance. Colloidal solution of
noble (nonoxidizable) metals, of many metallic
oxides, sulfides, and "insoluble salts" are largely cov-
ered. They show themselves optically heterogeneous
by Tyndall beam and ultramicroscope, and their be-
havior is largely representable by supporting the
idea of mechanical subdivision with that of specific
adsorption of electrically charged "ions" to their
surface, giving them an electric charge opposite to
that of the medium. But, it is precisely for the emul-
soid colloids primarily considered by Graham —
gelatin, albumin, globulin, rubber — the colloids par
excellence — that the conception just outlined appears
inadequate. Their properties and behavior appear
better explainable on a development of Graham's
original conception. Many of these emulsoids, when
carefully freed from electrolytes, show only the faintest
traces of optical discontinuity. The facts point to
their solutions being crystalloid in point of "dispersity,"
while their behavior to acids, alkalies, and salts is
best explained in terms of definite chemical reactions.
Their outstanding physical property, of forming very
viscous solutions readily passing to elastic gels, is ex-
plicable by the formation of tenuous networks, of
molecular and submolecular mesh, woven perhaps by
the idiochemical affinity of Graham.
The true colloids do, however, pass by easy transi-
tions into the pseudo-colloids, for which the behavior
is less dependent upon the chemical character of the
molecules than on dispersity of mass.
Although emulsoids might be supposed more kin
to emulsions than suspensoids, yet an emulsion is a
good model of a suspensoid. Hence, all in all, I
think we may say that the development of colloid
chemistry has been perfectly paradoxical. Like the
completely irregular Brownian movement, which has
formed a focus of certain aspects of colloid science,
it is impossible to fix even approximately a tangent
at any point of the trajectory of any particular develop-
ment of the science. And this atmosphere of unlimited
possibilities lends a fascination to what at first seems
a repellent medley of empiricism and speculation.
COLLOIDALIZING FUELS
In considering the problem of stabilizing a suspen-
sion of coal or other carbonaceous matter in oil we
can best start from a mathematical law for the fall of
bodies in a viscous medium, i. e., one offering resistance
to shearing. Stokes' law states that the steady
velocity of fall of a spherical body is given by the
formula:
y 3r»(S~S')g
go
where r = radius of particle
S = specific gravity of sphere
S' = specific gravity of fluid
g = acceleration per unit mass (gravity)
v = absolute viscosity of fluid
The pulverized coal first tried was a semi-anthracite
of sp. gr. 1.467; the specific gravity of the oil was
°-8oo7(2o-°). its absolute viscosity 6. The radius of
the coal particles could be taken as a first approxi-
mation as one-half the aperture of the screen they
passed through, or one-quarter of the reciprocal
of the mesh number. From these conditions we
should have had:
Mesh to Which .
Coal Was 2r Calci
Pulverized Cin. In. pi
50 0.0127 9
100 0.00635 1
200 0.00317
400 0.00158
-Rate of Fall
Actual
Inappreciable i
Appreciable in 4 weeks
The coal used was between ioo and 200 mesh fineness,
and there was about 30 per cent by weight present.
The wide deviation from Stokes' law was in the right
direction and so far promising. It could be tentatively
explained:
1 — -By nonspherical form of the particles. As platelets or
spicules they would not fall straight.
2 — By increased inner friction or mutual impedance in the
concentrated suspension. However, "clumping" would accel-
ii. id settling.
.; -By some kind of combination, e. g., capillary adsorption,
with the oil.
The oil first used was moreover a nondescript ma-
terial, very viscous — though not so viscous as Mexican
fuel oil. It so happened that the first supplies of oil
now brought for trial were either Texas Oil Company's
Naval Fuel oils, of relatively low viscosity (around
200 Engler) or Standard Oil Company's Naval Fuel
oils, of even lower viscosity. We soon found that
fuel oil is a very variable material. It is well known
that mineral oils vary greatly in chemical composition.
While Pennsylvania oils, of so-called paraffin base,
do contain considerable proportions of saturated open-
chain hydrocarbons, together with lower members of
the cyclic olefines, the midcontinental American oils
have more of the cyclic olefines, also asphaltie hydro-
carbons (malthenes, carbenes, etc.) and "free" carbon.
More important for present considerations is their
great variation in physical properties. Fuel oil
is a residual product, left by removal of the lighter
fractions suitable for gasoline, kerosene, etc., and now
still further diminished by various cracking processes.
The oil refiner grades his oils chiefly by gravity.
Expressed in terms of the Baum6 scale, they show
pretty wide variation, yet in terms of specific gravity
it is not so considerable. For the problem of stably
dispersing coal or carbon in oil, the variation of grav-
ity, from 0.85 to 0.96, is not so formidable as the range
of viscosity. This can and does vary from 1 to 30,000,
in terms of specific viscosity of water. Again, this
viscosity varies greatly with temperature.
In our first work, as stated, we encountered the thin
end of the wedge with oil of about 20 ° Engler. It
was not found possible to prepare stable composites
with this oil untreated, even with coal pulverized so
that 99 per cent passed 200 mesh. To discuss the
actual stages of treatment as the problem presented
itself would take too much time and space. It was
evidentthat it was necessary:
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. r
BHVJ t>90 ' 3UniYHidH2X
I — To find working standards for the minimum and maximum y — To find protective colloids adequately stabilizing the com-
viscosity permissible of the oil base. posite within permissible viscosity limits.
2— To approach the practicable viscosity minima of stable There are other factors, to be touched upon, but
composites to specification maxima for atomizable fuels. these three are dominant. Yet they are very closely
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
41
interwoven, and interdependent. First, they have
to be considered in regard to temperature. The
viscosity of fuel oils sinks rapidly with rising tempera-
ture, as shown in the diagrams (Fig. 1).
This has to be considered in relation to flash point.
It has been found1 that for effective atomization by
mechanical burners the viscosity should be reduced,
by preheating, to about 8° Engler. Greater reduction
gives no marked advantage. To secure this, the
temperature to which the oil may be heated must not
be higher than its flash point.
\
BINE
ER
- R
5. 01
,\
5%
>-
5i
•ox
«
^
A= STA
IDA
m
JIL
5%
p%
""""~
TO
1
0
131
Temperature Fahrenheit
Fig. 2 — Blended Oil Curves
We have then one terminal pair of values to be
worked to:
Viscosity, 8o° E.; Temperature, flash point
British naval specifications for the flash point were:
not lower than 1750 F. closed cup, or 2000 F. open
cup; U. S. A. specifications: 1500 F. closed cup, or 175°
F. open cup. Considering then, for the original pur-
pose, that a close approximation to naval standards
was desirable, it had to be aimed to make the terminal
pair of values of the viscosity-temperature curve of
the composite fuels 8° Engler at 150° F. There is,
however, evidently a certain latitude, in that with
higher flash points a higher preheating temperature for
the same viscosity is permissible. Again, the viscosity
depends upon the pressure of injection.
While blending at first was mainly a problem of
thickening thin oils to suitable minimum viscosity to
permit of practicable amounts of the "stabilizer"
or "fixateur" being used, it later became rather a
question of suitable maximum viscosity, so that too
thick a fuel did not result. It might be thought that
this latter condition simplifies the stabilizing problem,
in so far as stability depends upon viscosity. This
is partly true, but not entirely. In very viscous
fuel oils, such as Mexican Panuco, etc., there is a strong
tendency for "free carbon" and suspended carbon to
clot. So that there also the role of "protective
colloids" as also of peptizers and deflocculators is very
important. Before passing to these aspects, let me
point out in conclusion of this section that "blending"
meant adjusting the oil base to a standard viscosity-
temperature curve (Fig. 2).
So great are the varieties of these curves with differ-
ent materials, and so large the deviation from any law
1 E. H. Peabody, "Oil Fuel," Trans Internal. Eng. Cong., 1915.
of mixtures — whether for viscosities or fluidities —
that this has to be done by "trial and error" methods
in the main.1 But, technically, it has been adequately
solved, and a great amount of valuable data secured.
Commercially, it is subject to local and temporal
conditions of availability.
STABILIZATION AND PROTECTIVE COLLOIDS
As already stated, the problem of stabilizing sus-
pensions of carbon in oil is not solely one of getting
viscosity in the oil medium. While heavy paraffins
and cyclic defines give viscosity — and have also
much protective value as semicolloids themselves —
they are too valuable, as lubricants, to be very avail-
able in fuel oil. The more viscous residuals available
for increasing viscosity are asphaltic materials, con-
taining large amounts of "free carbon," in colloidal
suspension, but tending itself to clot and settle out.
There are two ways of stabilizing this, of which the
first we need consider is the use of protective colloids.
Protective colloids in aqueous systems are well known,
e. g., gum arabic, gelatin, glue, etc. They are classed
as emulsoids, or lyophile colloids — the first name
from the idea that they form a submicroscopic liquid
dispersed phase, the second from their affinity for the
solvent. It is the second conception which is the
/
,
/
/
/
/
,
/
1
11
/
1
/
[
/
1
.
//
1
'/
1
Q
/
!
S
7
g M
/
/
/
/
o
7
1
It
/
/
1
o
4
'
r
,"/
I
C
A
1
1
/
iJ
;
/
f
s
-
j
/
/
i
/
/
i
/
/
,,'
&
a?
i
'
/
i
'
i
_!
„
r
1 —
m
PERCENTAGE
Fio. 3 — Curves Showing Viscosity op Fixated Oil in Relation to
Concentration and State of Protective Colloid
more important. Substances forming emulsoid col-
loids in nonaqueous media are also known. Many
1 See the recent and valuable paper by W. H. Herschel, "Saybolt
Viscosity of Blends," Bureau of Standards, Technologic Paper 164 (1920).
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. i3> No.
soaps, particularly of the alkaline earth metals, such as
lime soaps, form emulsoid colloids with mineral oils.
It is a group of these which furnished the fixateur, or
protective colloid used to stabilize suspended carbon
in colloidal fuel.1 Like emulsoid colloids in water, the
preparations of these soaps in oil show a very rapid in-
crease in the viscosity with increasing concentration of
colloid (Fig. 3).
This viscosity-concentration curve is very irqportant
in judging the adequacy of dispersion of the colloid,
on the one hand, and the measure of its protective
action on the other. With the particular type of emul-
soids we have to deal with, the steepness of this curve
depends markedly on the mode of preparation. It
appears that every gradation exists between the mark-
edly emulsoid condition and suspensoid dispersion,
in which the system is much less stable.
QUANTITY OF FIXATEUR AND VISCOSITY
The amount of fixateur which could be used was
approximately fixed by conditions of cost, and varied
from 0.5 to 1. s per cent. Although the immediate
effect is to thicken the oil, i. c, increase its viscosity,
it is to be remarked that increase of viscosity alone is
not the sole condition conferring stability of suspension
of carbon or pulverized coal, coke, etc. Oils thickened
by other means, e. g., by vaseline, to the same viscosity,
gave much lower stabilities. It was repeatedly
found that viscosity, while an important factor, was
not the only one. This is already known to be the
case for the protective action of emulsoids on suspen-
soid colloids, and evidently extends to suspensions.
PLASTIC INNER FRICTION
On the whole, it is probable that the immediate
condition for protective action is strong adsorption of
the colloid to suspensoid or suspension. But this
does not entirely account for the mechanism of pro-
tection. ! believe we may account for this by the
tendency of these colloids to form heat reversible
gels. Such gels — not coagula — may be imagined as
very tenuous web-work or foams, the mesh or walls
1 The Submarine Defe e Association, a war organization, dissolved
and terminated its existence at the close of hostilities. During the w.ir
it sponsored the new fuel. All patents, trade-marks, copyright and other
rights in the fuel are in Mr. Lindon W. Bates' name and are vested in a
company. Release of patents since September 1920 has allowed explicit
statement of the fixateur to be made.
of which are very probably submolecular in dimensions;
or, if we like, the whole mass of colloid forms one
"molecule" uniformly dispersed through and partially
dissolving the solvent. By partially, I mean that part
only of the "molecule" of the emulsoid is consolute
with the solvent or dispergent, while the other part
of it is insoluble, and its atoms tend to unite, forming
a semirigid framework. Such a system would have
the following properties, which are observed in jellies:
1 — Offer little resistance, unless very concentrated, to diffu-
sion of solute.
2 — Offer little resistance to powerful shearing stress, or move-
ment of heavy bodies.
3 — Offer great resistance to very small shearing stress, or move-
ment of very small masses.
That is, such systems would behave as fluids for in-
ternal diffusion of solutes, and for shearing stress of
appreciable magnitudes, but approach the behavior
of elastic solids for internal movements of small magni-
tude. Internal friction of this type has been termed
"plastic," and is illustrated diagrammatically in Fig. 4.
Differential resistance of the kind noted is charac-
teristic of the plasmas or body fluids of organisms, and
it is such a plasma which is required for colloidal
fuel. Hence, it has really more than one coefficient
of inner friction, and the gross viscosity is not a com-
plete exponent of its inner state.
PEPTIZATION AND COLLOIDAL FUELS
I have said that there is a second method of im-
proving the stability of suspensoids and suspensions
of carbon in oils, other than the use of emulsoids or
protectives. This consists in peptization. The two
methods are probably connected. Protective action
probably means strong adsorption, and adsorption
leads to peptization. But it may not go so far. Pep-
tization for stabilizing graphite was employed by
Acheson, who used tannic acid as a defiocculator.
It was found in the present work that "free carbon"
in residual oils, such as pressure still oil and Mexican
oils, could be peptized and stabilized by addition of
certain by-products and distillates.1 This occurred
with a lowering of the total viscosity, due to the pre-
vention of clumping. Next, a still more remarkable
peptizing action of this type has been observed. This
was discovered as follows: We had found that the
peptizing of "free carbon" in petroleum residuals
could be extended to the problem of stabilizing dehy-
drated coal tars in mineral oil. Further, reasoning
by analogy with Pickering's emulsions, in which a
finely divided solid was found to stabilize an emulsion
of two immiscible liquids (oil and water), an attempt
was made to stabilize coal tar in oil by further addition
of pulverized coal. This attempt was largely suc-
cessful, a stability extending into weeks being secured.
We further added small amounts of peptizing substances
to these composites. On measuring the viscosity-
temperature curve of these, it was observed that when
maintained some time at relatively high temperatures
the viscosity, instead of diminishing, actually increased.
This thickening action was observed in detail. Dilu-
tion with xylene and microscopic examination, with
1 Notably creosote and naphthalene containing oils from tar.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
43
counting chamber, showed that the number of very
small to ultramicroscopic particles was greatly in-
creased, these showing lively Brownian movement.
Peptization or partial solution of coals by such means
is to be expected. The investigations of Bone, Wheeler
and others1 have shown that in general we may regard
coal as composed of three principal fractions, a, 0,
and y. Of these the a-portion is composed of com-
pounds insoluble in pyridine; the /3-portion is soluble
in pyridine but insoluble in chloroform; while the 7-,
--
P. S. O 1 u
— ■
'So '/too '/zoo /400 '/aoo '/1600 'Azoo '/6400 VlZBOO
Diameter of Particles inTrhctions of An Inch
Fig. 5 — Effect of Subdivision of Coal on Viscosity of Fuel
or resinic portion, is soluble both in pyridine and chloro-
form. It is well known that the oils distilled from
resinous bodies such as amber, copals, rosin, rubber,
etc., are solvents for these substances themselves,
the solutions, however, being generally incomplete
(peptization). The microscopic examination of coals1
tends to show that with certain exceptions coal is far
from being a physically or mechanically homogeneous
material, resultant of pyrogenic metamorphosis. To
quote Wheeler and Stopes:2
We conclude that coal is a conglomerate of morphological
organized plant tissues, natural plant substances devoid of
morphological organization (such, for instance, as resins) together
with the degradation products of a portion of the plant tissues
and cell contents comminuted, morphologically disorganized,
or present in the form of varying members of the ulmin group.
From this it will be seen that the efficiency of pep-
tization by tars and distillates is likely to vary con-
siderably from one coal to another, and again to some
extent with different particles of the same pulverized
coal. In practice, this is found to be the case. Ac-
tually, however, cannel, bituminous, and even an-
thracite coal have been found peptizable by these
methods. Such peptization does not, alone, neces-
sarily produce complete stabilization in the oil-tar
medium. Generally it is easy to secure 3 to 4
wks. of homogeneity. After this the composite
gradually separates into an oily supernatant top layer
over a more viscous mass. This lower layer, however,
is usually quite easily remixed, and only very slowly,
if at all, tends to pass to a dense, solid mass. Usually
the lower stratum forms a more or less mobile jelly,
1 M. C. Stopes and R. V. Wheeler, monograph on the "Constitution
of Coal," Department of Scientific and Industrial Research of Gt. Britain,
London, 1918.
J Loc. cit.
showing synaeresis, *. e., shrinkage, with exudation of
oil. We have provisionally termed these the B-type
colloidal fuels. They are, per se, more readily and
cheaply compounded than the A-type, in which
stabilization is effected by an external protective
colloid — the fixateur — and are perfectly satisfactory
as liquid fuels for land installations. Finally, processes
of this B-type may be combined with those of the A-
type.
limits of peptization — The peptization process,
as stated, increases the viscosity. This may be
partly due to an extraction of "resinoid" bodies, but
no doubt is also due to increased dispersity of the coal.
For, as the dispersity of a suspension is increased,
the viscosity, or rather the inner friction, is also. This
is illustrated in the diagram in Fig. 5, for the case of a
30 per cent coal suspension. It is evident that pep-
tization must not be pushed too far, to excessive vis-
cosity.
alternative methods of peptization — An alter-
native method of peptization involves an entirely
different method of attack, viz., attack on the cellulose
and "fixed carbon" portion by oxidative reagents,
either wet, or gaseous. Anthracite coals contain a
very condensed cellulose fraction which approaches
free carbon in behavior. Carbon and coal both yield
mellitic or graphitic acid (benzene hexacarboxylic
acid?) on oxidation. Partial oxidative attack need
be relatively slight, in percentage oxidation, while
giving considerable peptization, and this method is
also available for the production of colloidal fuels.
Hence, we have three methods or. stages of attack,
resulting in progressively more deep-seated attack:
Mechanical Solvent Chemical
Comminution Peptization Peptization
Fig. 6 — Rocking Storace Tank Sb
ig Two Positions
ACCESSORY TESTING METHODS
Just as the proof of a pudding is in the eating, so
the tests of a fuel are essentially keeping powers and
combustion efficiency. Of these I will speak directly.
But in the technologic development of these fuel?
various laboratory accessory tests were devised. It
has been stated that fuel oil on shipboard tends to
separate water which is not separated on land storage.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
Fig. 7 — Capillarimeter
The fuels were therefore tested for such "seasickness"
in the apparatus shown in Fig. 6, which has a motion
approximating the pitching and heaving of a vessel,
and no difference .was observed. Rapid methods of
analysis for the free carbon in suspension were devised,
including a centrifuge for washing out the carbon
while running. Further, rapid centrifugal and capil-
lary methods of proximate stability testing were de-
vised. By these a partial prediction of the life of a
fuel is possible. The accelerated test by centrifuge
consists in determining the force required to effect a
given per cent separation, and this is calibrated on
5
$
t/
1 *
r
c
oj2-
??'-
—
\~
/
,
s
'
y
'
0
I
'
0
1
i
gravity stability trials. I say calibrated, because a
direct relationship does not exist here. The capillary
method is based on this. Oil plus fixateur plus carbon
are held by capillary chemical attraction, at the least.
If we put in a piece of standard porous paper, the oil
will climb this the faster, the less it is held back by the
combination (Figs. 7 and 8).
Further, it was necessary to determine the viscosity-
temperature curves of base oils, fixated oils, and com-
plete fuel. For proximate work, a pipet of special
type, running as many seconds as degrees Engler, was
used, as well as Engler and other viscosimeters. Other
essential determinations, on raw materials, inter-
mediate stages, and completed fuels, were specific
gravity, B. t. u., ash, sulfur, moisture, etc., also flash
points, and ignition temperatures.
METHOD OF COMPOUNDING
The machinery for compounding these fuels is
simple. It consists of a suitable mill for pulverizing
coal, coke, etc., storage and blending tanks for the
M/NUTES
-Showing Capillary Rise with On. and Fuel, Respectively
Fig. 9 — Cost Chart. Reproduced from a Pamphlet on "Colloidal.
Fuels, Properties, Tests and Costs," by Lindon W. Bates, 62 Lon-
don Wall. London, England
oil bases, and mixing kettles for compounding the
composite fuel. Little modification in existing types
of machinery is necessary, and the process is readily
made continuous. The cost of manufacture may be
reckoned at approximately $1.50 per ton, inclusive of
fixateur. The general relation to cost of oil is shown
in Fig. 9.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
PROPERTIES OF COLLOIDAL FUELS
It will be evident that the colloidalizing process is a
flexible one, allowing a great number of grades and
varieties to be produced. Standardization of grades
has already been commenced, but the flexibility possible
is valuable, in view of adjustment of the process to
local or temporary conditions of supply and demand.
1 U.J.Jnl Mil
|Cu.frJ | | | |
' " ." ' \\\\\\\\\\\^
1 1 1 1 1 1
k\\\\\\\\>
0.7MC.n OiL.0.4»«C»Fr fe«>ii»CoAi..ftKUCAr< R™cr«.ll»t
1 1 o'.L 1 1 1 1 fM^mU
1 1 1 1 1 1 1 1 1 1 | 1 M 1 1
07«Cuft0ii. OJ«J£«lt».-.»K..Co»L > - - 19.01
1 ln.1 1 1 1 1
1 1 1 1 1 1 1 1 1 M II
| V«.UMC or Oil »,IK 1«B6 WT. «S ICKrtCGU.MM.KW.
I I 1 1 I p;LVEaiieGCo«L 1 1 1 1 1
II II M 1 1
^^W^^^^^
^\^\N
1 1 jiygjjjjyiU 1
1 1 1
'II II
TTl"!TtTTiTi rr
| VoLO«toro.LB«v,NOSA«IHiAtu«,rjA5Uo.rT.coLLOio»i.F»Gu 1
Chart Comparing Volume of Coiloidal Fuel wit*
Aggregate Volumes of Coal andOil, and
•• * Oil = 16*69 " " Cool' i.i
K
0
Coo-
so
0.2 04 0.6 0.8 1.0 1.2
1.6 16 10
Fig. 10 — Volumetric Comparisons between Oil, Coal, and Colloidal
Fuel
This table shows graphically the volume occupied bv colloidal fuel after
manufacture; the volume before colloiding; and the volume of oil with the
same weight as a cubic foot of colloidal. 1.02 cu. ft. of oil have the same
heat units as 1.0 cu. ft. of colloidal fuel, which shows a gain in cruising
radius per unit of space for colloidal. The chart also shows that pulverized
coal is nearly twice as bulky as colloidal fuel for the same number of heat
units. To illustrate, if a high-grade navy oil is used with high-grade
Cardiff coal, we obtain the most compact fuel known per unit of space.
Meanwhile, the following brief summary of the proper-
ties of colloidal fuels is in order:
(1) They are liquid, and handle and atomize for combustion
like fuel oil.
(2) They can be made to contain more heat units per gallon
than fuel oils. This is a consequence of the law of mixtures.
The specific volume of the colloidal fuels is lower than that of
the oils they are made from. Fig. 10 shows graphically the rela-
tion of heat units to volume. In general, they will weigh from
8.75 lbs. to 1 1.5 lbs. per gal., according to kind and per cent
of carbon, e. g., coke, coal, pitch, or lignite, employed.
(3) They contain very little moisture and ash. The ash
obviously depends upon the kind and per cent of carbon incor-
porated, and can be kept very low by use of high-grade carbons
or de-ashed coals.
(4) Flash point is above 2000 F. They are immune from
spontaneous combustion. The so-called spontaneous combustion
of coal in piles, bunkers, and as powdered coal is due to initial
fixation of oxygen of the air. self-heat, and autocatalyzed autox-
idation.1 Immersion of the coal in oil prevents the first step,
the formation of addition complexes of oxygen and coal compo-
nents.
(5) Not only are they vaporless up to high temperatures, thus
avoiding explosive mixtures with air, but they may be fire-
proofed by a "water seal" of an inch or more of water, due to their
specific gravity being higher.
(6) Hence also they will sink if spilled blazing on the surface
of water, i. e., are self-quenching. They are quenchable by
water with ordinary fire apparatus where the surface may be
covered, as also by sand, Foamite, etc.
Summarizing their safety factors, their fire-risk is as low as
anthracite coal, and far safer than bituminous coal or ordinary
1 Porter and Ralston, "Study of the Oxidation of Coal," U. S. Bureau
of Mines, Technical Paper 65 (1914); R. B. Wheeler, "Oxidation and Igni-
tion of Coal," J. Chem. Sue., 113 (1918). 945.
fuel oil. These properties have been investigated by the National
Board of Fire Underwriters' Laboratory. They have substan-
tially confirmed them, and reported to the Fire Council that all
installation using colloidal fuel be given the benefit of standard
fire rates. The Council adopted the recommendations.
(7) Storage Test — They are the most compact fuels known.
A cubic foot contains 7.4805 U. S. gal. An average bituminous
colloidal grade contains 160,000 B. t. u. per gal., or 1,169,800
B. t. u. per cu. ft. With anthracites and cokes up to 1,346,490
B. t. u. per cu. ft. may be realized. The advantages of this are
obvious: increased radius for ships, and lessened storage space
in crowded cities.
COMBUSTION EFFICIENCY
The following table shows what was accomplished,
first with straight A-type fuel, stable for 6 mo. in marine
trials, and secondly, with A- and B-type fuels on land.
Table I — Typical Result of Steam Tests on U. S. S. Gem, S. P. 41, 1918
Fuel , : Colloidal . . Navy Oil .
System Standard Schutte & Koerting Mech. 1.7 Mm. Burners
Test number 2 4 6-B 12 3 6- A
Date April 18 April 30 May 3 June 22 April 19 May 3
Duration 2 hrs. 2.25hrs. 0.67 hr. 3.17 hr. 2 hrs. 1.5 hr.
Feed water temp
entering heater 72.1 83.3 81.0 95.3 77.5 71.3
Feed water temp.
entering boiler. 233 195.3 229 218 223 177
Flue gas temp,,
average 745.5 668.5 644.5 629.5 661.5
Air temperature,
outside 45.2 60.6 72 60.5'
Air temperature,
boiler room 66.7 75.6 65.8
Air temperature,
engine room. . . 81.0 80.0 68 82.2 .... 67
Fuel temperature 173.5 159.8 155 139 134.3 140
Fuel pressure, lbs. 131.8 149.5 156.5 101 96.1 125
Draft uptake in
WG 0.05 0.05 0.05 0.05 0.05 0.05
Draft pressure,
wind box, in
W. G 0.73 0.71 0.70 1.08 0.72 0.66
Vacuum 25.0 24.8 25.8 24.7 26.0 25.3
Barometer 29.88 29.91 25.90 29.71 30.35 29.90
Smoke average... 30% 0-10% 0-10% 0-10% 10% 0-10%
CO; 8.5 7.0 11.2 8.6
Boiler pressure,
lb. g 208.4 249 240 232 235 220
Engine pressure,
lb. g., average. 73.2 86.2 117.5 119 90.8 85.45
Intermediate pres-
sure, lb. g., av-
erage 17.35 22.9 35 35.4 25.0 22.8
R. P. M 214.5 231.6 260.5 268.2 243.5 230.5
I. H. P. main en-
gines 439 514 677.6 851.4 569 530.4
Knots by Log.... 11 13.64 14.64 12.25
Knots by Sanborn
gage 11 13.4 14.18 13.3
Fuel per hr, lbs.. 950 1030 1050 1493 938 1000
Assumed water
rate. lbs. per I.
H. P
Steam per hr.
from and at
212° F 10970
Factor of evapora-
tion 1.029
Evaporation per
hr 10660
Lbs. water per lb.
fuel 11.25 11.65 15.6
B. t. u. per lb. fuel 17100 17100 17100
Evaporative effi-
ciency, per cent 65.5 70.7 91.5
Lbs. water per sq.
ft. heating sur-
face 4.06 4.56 6.23 8.03 5.19 4.64
Lbs. fuel per I. H.
P. main en-
gines 2.16 2.0 1.55 1.75 1.65 1.9
gem tests — In this first trial under service condi-
tions, the fuel consisted of 31.2 per cent Pocahontas
coal of 13,974 B. t. u. and 67.8 per cent Texas fuel
oil of 18,669 B. t. u. The B. t. u. of the composite
was 17,100 per lb. 1 per cent fixateur. With fuel 3
to 4 mo. old, tests were made on Long Island Sound,
directed by H. O'Neill, then engineer of the West
Virginia Pulp and Paper Company. They were
witnessed by representatives of the American and
Allied navies, the U. S. Shipping Board, and other
fuel experts.
24.3
23.5
24.2
24.0
12850 16950 21300 14200 13250
1.071 1.035 1.048 1.039 1.088
12000 16370 21100 13660 12200
79.4 69.6
4«
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
Table II— Da
and Results op Boiler Tests of Colloidal Fuel, 1919
Grate surface, sq. ft
Total heating surface, sq. ft
Date
Duration, hrs
Kind of liquid colloidal fuel, grade. .
Steam pressure by gage, lbs. per sq. in
Temperature of feed water entering boiler, deg
Percentage of moisture in steam or number of degrees of super-
heating, per cent or deg
Percentage of moisture in liquid colloidal fuel, per cent
Liquid colloidal fuel per hour, lb
Liquid fuel per sq. ft. grate surface per hour
Equivalent evap. per hour from and at 212°, lb
Equivalent evap. per hour from and at 212° per sq. ft. heating sur-
face, lb
Rated capacity per hour from and at 2 1 2°, lb
Percentage of rated capacity developed, per cent
Equivalent evap. from and at 212° per lb. of dry coal, lb
Equivalent evap. from and at 212° per lb. of combustible, lb
Calorific value of 1 lb. of fuel by calorimeter, B. t. u
Calorific value of 1 lb. of combustible by calorimeter, B t u
Efficiency of boiler, furnace, and grate, per cent
Efficiency based on combustible, per cent
3.61
403 b. h
126%
13.6
2.8
1076
i5942
3.29
403 b. h. p.
115%
14.72
16670'
85.3
1159.5
16200
403 '
110.6%
13.97
18482
-'3.3'
1146.5
17202
403
1221 ;
14.85
.94
1054.7
16567
403 '
118.8%
15.51
18482
79! 46
982.75
146i2
403"
105%
14.89
18482
7o!s'
Compositions — Colloidal Fuels
Grade
1 1
Per cent
Numbers
14
Per cent
Coal
Coal (Pocahontas).
Coal tar, etc
Fixateur
Mexican reduced.
Pressure still oil
The second series of trials took place on land at the
Standard Oil Refinery in Brooklyn. The boilers used
were old type tubular return, 5 to 7 per cent less efficient
than later B. & W. or Sterling types. Fuels of both
A- and B-types, and mixed grades were used; the
"peptization" process fuels were burned with complete
Fuel Oil
Floating on tyah
Colloia Colloidal Fuel
Sealed under Wafer Kepi 1 ueor under Water
Average efficiency, 76.37 per cent
Analysis Grade 13
Ash 3.20 per cent
Sulfur 1 . 27 per cent
Viscosity, 70° F 67.5° Engler
Sp. Gr , 70° F 1.0431
Flash 250° F.
Fire 285° F.
Moisture 0.2% per cent
Grade 14
Ash 2 per cent Sulfur 0.2 per cent
in the Bone-Court flameless superficial combustion
procedure. Now the atomized coal plus ash particles
provides an enormous internal surface. The fume
of partly burnt coal and ash particles in the combustion
space gives an added surface factor, which, under
proper conditions, makes the efficiency of these fuels
equal to or greater than that of higher grade straight
oils, having no solid particles present. Further, with
increased percentage of carbon there is less heat loss
by steam formation.1
1 As stated, the liquid types of colloidal fuel require no
special arrangements for burning, either air, steam
TIME IN MONTHS
I 9 ■* A. 5 £ 7 a
100 r
1 1
Typicm. "Life' Curves
*
s
1
1
1
1
1
i
/
/
/
1
*
1
1
>
-
/j
A
/
/
*"''
v"
1
Time in Months
-Life Curves of Colloidal Fuel — Show Prolongation
BY REAGITATING AFTER INITIAL SETTING
Excellent results were obtained with other fuels,
using 40 per cent anthracite rice (pulverized) contain-
ing 25 per cent ash. The remarkable fact that these
fuels, actually of lower grade than straight fuel oil. in
B. t. u. per lb., are capable of giving equal or higher
boiler efficiencies is, we believe, explained by the follow-
ing considerations. Combustion efficiency of oils and
gases is greatly increased by surface. This is shown
injection, or mechanical burners being suitable. It
is desirable to have a steam by-pass on the burners to
"blow-through" after turning down.
"life" curves
Finally, what is the period over which those fuels
can be made intrinsically stable? "Can" and "need"
1 The calorimetry and heat balance of these fuels will be discussed
more fully by Mr. H. O'Neill shortly.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
must be distinguished here. I believe that they can
be made just as stable as needed. Samples prepared
in the laboratory have lasted 12 to 18 mo. in quite
stable form (Fig. 12).
An important point here is that reagitation, before
sedimenting has progressed too far, will give a further
extension of life. The grade of fuel can be fitted to
the conditions of permanence and stability required,
and is technologically related to the dispersity gradient,
the varying properties of particles of different dimen-
sions in it. Colloidal fuel is a composite dispersoid,
the particles of which range from solution through the
colloid to suspensions. With every advance in the
technique of the subject, the right proportioning and
grading of those for a given purpose becomes better
understood, and the relation of the dispersity gradient
to stability and use becomes clearer.
FUEL CONSERVATION, PRESENT AND FUTURE
By Horace C. Porter
1833 Chestnut Street, Philadelphia, Pa.
Progress in the application of fuel to the needs of
mankind is being manifested in an improvement of
methods, a rise in the curve of efficiency, as well as
in that of total consumption. To-day, resulting from
increased use of scientific methods, we see greater
returns per ton of coal than 10 yrs. ago.
The per capita consumption of fuel in the United
States has increased by only 7.5 per cent in the last
10 yrs. — from 152.3 to 163.9 millions of B. t. u.
The increase has been in oil and gas, not coal. It
is cause for congratulation, therefore, that notwith-
standing greater industrialization, higher standards
of living, and the devoting of vastly increased indus-
trial yields to the benefit of other nations and of our-
selves, we have maintained so small an increase in
fuel consumption.
Fuel production is with difficulty, however, keeping
up to the demand. Under the trying conditions of
the last few years, transportation deficiency has
retarded fuel distribution and production, so that a
real shortage exists to-day. The loss of 50,000,000
tons from the normal coal production during the
nation-wide coal strike of 19 10 put industry in the
position of holding back needed improvements and
new construction which now are calling urgently for
more fuel. Stocks also need to be built up. Exports
from tidewater have leaped to 600 per cent in 2 yrs.,
and threaten to pass 25,000,000 tons for this year.
In the face of these facts, and of the impression
prevailing in many quarters of a dwindling coal pro-
duction, it is in a measure reassuring to note that for
the first 6 mo. of this year coal production is 19 per
cent greater than in the corresponding period of
last year, and oil is 15 per cent greater. As compared
similarly to 191 7 and 19 18, war years, coal has this
year fallen behind by 5 and 10 per cent, respectively.
Reconstruction now urges upon us the use of addi-
tional fuel. To emerge from the transition period
of 1919 and make this truly a reconstruction year,
our industries must be given the necessary coal and
oil. As to how far we fall short now of our proper
share in the world's reconstruction, the economists
can perhaps make better guesses than chemists and
engineers. But in point of coal consumption we may
make comparison with 1918 when expanded war in-
dustries brought this item to the highest point it has
ever reached in this country, before or since, and
find that our present rate is but 10 per cent in arrears,
of which probably half can be accounted for by increase
in exports.
Professionally, to the industrial chemist and engi-
neer, conservation appeals as an important aid in
removing or reducing fuel shortage. A reasonable
and practicable increase in fuel economy would help
materially in bringing supply and demand closer
together. There would be exerted in consequence of
it, also, an influence toward lowering of prices. No-
table advance has been made during recent years,
but the practical maximum of efficiency has by no
means been reached. There is not to be overlooked
or minimized the tendency of human nature to use
available natural resources to the limit, with little
regard for posterity. Yet in times of shortage in
supply, the consumer perhaps has his interest more
easily aroused in means of cutting down requirements
and reducing raw material costs.
Bituminous
Coal Used Per cent
(Net Tons) of Total
Possible Means of Conservation 1917 Consumption
(1) Industrial Power 130.150.000 23.4
(excl. steel mills and coking)
(a) Increased use of economizers,
superheaters, feed-water heaters, me-
chanical stoking
(6) Care in firing, with control of
flue-gas composition and temperature
(c) Use of gas engines in conjunc-
tion with steam, on power plants
where load is variable
(2) Steel and Iron Industry 90.000,000 16.2
(excl. coking)
(a) Increased use of gas for heating
and power, and of regeneration and
recuperation
(6) Increased use of waste heat
for steam generation
(c) Powdered coal and tar in heat-
ing furnaces
(3) Beehive Coking 52,250,000 9.4
(a) Gradual abandonment in favor
of by-product coking
(i>) Utilization of waste heat in
boiler firing
(4) By-product Coking 31 ,500,000 5.7
(a) Increased utilization of waste
heat through regeneration, recupera-
tion and steam generation; increase
in surplus gas and its utilization
(5) Railroads 156,150,000 28.0
(a) Use of feed-water heaters and
economizers on locomotives
(6) Economy of steam pressure by
idle locomotives
(6) Domestic 57,100,000 10.2
(a) Avoidance of unnecessary heat
in unused places and of excessive
temperature when not needed
(b) Economy of gas used as fuel by
adjustment of appliances
(7) Other Uses 39,700,000 7.1
(Gas manufacture, export, and bunk-
ering of vessels)
Total 556.850,000 100.0
Many of the expedients for raising the efficiency of
fuel utilization are of such a nature as to require large
changes of existing plant and equipment — the cen-
tralization of power development, for example, in
super-power stations, the electrification of railroads,
and the building of by-product recovery coke plants.
These changes go slowly, and depend greatly on general
financial conditions and the prevailing cost of capital
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. i3) No. i
•outlay. Other expedients afford in the meantime
quicker realization of efficiency gains, not as large,
but of distinct importance in practical conservation.
The preceding tabulation of the country's coal
consumption in 191 7, by classes of users, is taken from
the U. S. Geological Survey reports, and is coupled
with an outline of some of the means whereby con-
servation might be accomplished in the different
fields without great delay.
PRESENT CONDITIONS
It is to be noted that seven-tenths of all the coal
is burned under industrial and locomotive boilers and
in metallurgical heating furnaces. It is in this large
field that perhaps the most immediate opportunity
for improved efficiency exists.
boiler furnace EFFICIENCY — In boiler furnace
economy roughly half of the efficiency losses are due
to heat carried away in the chimney gases; under
commonly prevailing conditions an increase of 1 in
the percentage of C02 in the chimney gases means a
lowering of the excess air by about 10 per cent, a con-
sequent reduction in the B. t. u.'s carried away in
sensible heat, and a gain of 1.5 to 2 per cent in the
combined efficiency; a lowering of the flue-gas tempera-
ture by ioo° F. means an additional gain of over 3
per. cent in boiler and furnace efficiency. It is some-
what startling to those who have not stopped to con-
sider the matter carefully, to find that for every
pound of coal burned, 15 to 25 lbs. of chimney gases
result, carrying out their sensible heat to waste.
These efficiency gains are not in large figures, but
they mean a good deal when applied to the large ton-
nage of boiler fuel used.
Superheaters and feed-water heaters, if more gen-
erally applied, would add further to the saving. D. D.
Pendleton1 has recently estimated that only 15 per
cent of the steam raising capacity of the country
is equipped with superheat, and that the remainder
not so equipped would gain between 14 and 20 per
cent in efficiency by its use.
railway locomotive operation — In railway loco-
motive operation it is true that considerations other
than those of thermal efficiency are highly important
in obtaining the driving capacity required. On the
other hand, there are some opportunities for fuel
saving here, and it is a big field in point of total con-
sumption. In an article on "Locomotive Feed Water
Heating,"2 T. C. McBride has recently claimed that
devices for this purpose, utilizing the exhaust steam,
save on locomotives 10 to 13 per cent of the coal used,
as compared to injector operation. The maintaining
of high steam pressure unnecessarily in locomotives
standing idle in yards, the preventable part of the
so-called stand-by losses, is no doubt a factor in the
large railway consumption of coal.
industrial heating furnaces — A great deal of
coal is used in industrial heating furnaces for the
heat treatment and reworking of metals, the rolling
and forging of steel, and for tempering processes.
1 Blast Furnace and Steel Plant, 8 (1920). 350.
= Mech. Ens.. 42 (1920), 283.
Prof. H. M. Thornton1 has recently brought out the
great advantages and economy of gas as a fuel for
these furnaces. Records are presented showing com-
parative results in various sizes and types of furnaces
from the small rivet heaters to the large forging fur-
naces, the saving in fuel cost as compared to direct
coal, coke, and oil firing ranging from 40 to 60 per cent.
Indirect advantages also result in increased capacity
per unit and decreased labor cost. Prof. W. Trinks,2
of Pittsburgh, shows these economies in the use of
gas and of powdered coal in a series of articles on heat-
ing furnaces. The latter is pessimistic as to the
practicability of such savings, owing to the human
tendency of firemen to waste fuel when they can do
so easily by the turning of a valve. It would seem,
however, that under the inducements of a bonus
system this same ease of turning a valve might prove
a factor leading to conservation.
An actual record is given by A. A. Cole3 of a powdered
coal installation in a large heating furnace used in the
manufacture of rolled steel wheels, wherein an economy
of 30 to 40 per cent over direct hand firing was obtained,
and a labor saving equal to 1 5 per cent of the fuel cost.
At steel plants where by-product oven tar — an
excellent fuel for the open-hearth furnace — is available,
greater value frequently can be obtained from the
tar as fuel based on comparative coal cost at the
plant, than is obtainable in the open tar market.
Changes in open-hearth and heating furnace con-
struction designed to regulate combustion and length
of flame are proving in actual plant trials to effect
an increase in metal output, reduce waste heat losses,
and raise fuel economy by 10 per cent, without im-
pairing the life of the furnace.
waste heat boilers — More attention to waste
heat losses on industrial furnaces and in the older
by-product coke plants, with increased use, or more
efficient use of regeneration and recuperation would
pay well in fuel saved, giving added surplus gas at
the coke plants. Waste heat boilers are used on
many industrial gas-fired furnaces and by-product
coke plants. Their application could be widely ex-
tended with profit and an important degree of fuel
economy. Brick and pottery kilns, copper and zinc
and cement furnaces, and beehive coke ovens, show
waste gas temperatures from 12000 to 20000 F.
A large steel plant near Pittsburgh operates waste
heat boilers on the outlet flues of its rectangular non-
recovery coke ovens, obtaining thereby a steam output
which has reached 27 h. p. per oven. Reduced to the
basis of coal burned, this figure becomes in h. p.-hrs.
per pound of coal more than 25 per cent of the average
yield from complete combustion in steam plants.
miscellaneous — Large gas-engine-driven power
stations are being used by steel works on blast-
furnace gas with conspicuous success and large fuel
economy, as at Gary, Ind., by the U. S. Steel Corpo-
ration, and at Sparrows Point, Md., by the Bethlehem
Steel Corporation. Such means of power production
1 J. Roy. Soc. Arts, 68 (1920), 346.
* Blast Furnace and Steel Plant. 8 (192"
» Ibid. 8 (19J0), 417.
fan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
49
;an be extended, and a saving effected in coal more
than equivalent on a B. t. u. basis to the gas used,
swing to the comparatively high efficiency of the gas
;ngine.
Anthracite coal is being reclaimed from the river
bottoms in eastern Pennsylvania, and from the culm
banks by washing and briquetting. Culm also,
experimentally, has been mixed with pitch or bitu-
minous coal and carbonized.
In the domestic fuel field, comprising 10 per cent
Df the bituminous consumption (or 17 per cent based
Dn both anthracite and bituminous), the greatest
economies will eventually come from increased use of
jas and carbonized fuels. The domestic field will be
one of comparatively low efficiencies, however, as
long as small-sized fuel burning units remain. Econo-
mies can be made by using care as to overheating of
bouses, particularly of unused portions of houses.
Furthermore, in the burning of gas in domestic appli-
ances it has been shown by recent experiments at
Ohio State University1 that efficiency of utilization of
the heat may vary from 16 to 40 per cent, according
to the distances of the burner from the vessel heated.
FUTURE POSSIBILITIES
For the future, with the steady and permanent
growth of fuel economy through gradual adoption of
major improvements requiring time and large capital
outlay, there is reasonable prospect that the per capita
fuel consumption in this country may reach its peak
and begin to decrease, as in fact already the coal-
consumption curve, per capita, appears to have reached
almost its high level.
electrification of railroads — The most striking
possibility among these major improvements looking
to fuel conservation is the electrification of railroads.
It has been carefully figured by A. H. Armstrong, of
the General Electric Company, for the Committee
on Electrification of Steam Railroads, National Elec-
tric Light Association,2 that by universal electrification
of steam railroads in this country a direct saving of
122,500,000 tons of coal per annum, two-thirds of the
present railway fuel consumption, would result. This
leaves water power out of account and compares on
the basis of steam generated electric power in central
stations. Deduction is made from the present steam
engine ton-mile movement for company coal haulage
on cars and tenders.
The Chicago, Milwaukee and St. Paul Railway
has had in successful operation for over 4 yrs.
large electrified portions of its system in Montana and
Washington. The electrification now totals 645 route
miles. Power is purchased from the Montana Power
Company. In a detailed statement of actual operating
costs made to the^National Electric Light Association,
R. Beeuwkes, of the Milwaukee and St. Paul Company,
compares steam operated and electrically operated
divisions in respect to those items of expense affected
by the type of motive power used. For the totals of
these items electrical operation shows about 40 per
cent lower cost, and on the one item of train loco-
' Mich. Eng.. 42 (1920), 287.
» See Reports of this Committee, 1920.
motive power cost as against locomotive fuel used,
the saving amounts to 53 per cent, not taking into
account the cost of fuel haul.
These are direct savings, exclusive of the manifest
indirect advantages accruing from the release of freight
cars by gain in speed of haulage, the release to revenue-
bearing traffic of coal cars now hauling railway coal,
the avoidance of boiler feed-water expense, the im-
provement in reliability and safety of railway service,
and the increase of property valuation around railway
terminals. Most of these items will aid in decreasing
the menace of fuel shortage in the future.
High cost of installation, and the present difficulties
in the way of financing railway betterments, will act
to retard this great step in the progress of fuel con-
servation. The passage of the recent water power
legislation by Congress should, however, exert a large
influence in furthering such projects. Water power
development under favorable government regulation
not only affords low cost power, but releases coal car
equipment in greater measure than would central
steam stations. President A. H. Smith, of the New
York Central lines, has stated:
It is known that, generally speaking, the operating cost
(exclusive of fixed charges) of electric service is less than it would
be for a similar steam service; the further extension of
electric operation on steam railroads depends to a considerable
extent upon the cost of electric power; There is a point
where the cost of coal will cause the price at which electric power
is available to the railroad to result in sufficient saving to
warrant the expenditure for electrification.
centralization of power systems — The central
''super-power" station for general power service,
gradually displacing less efficient scattered units, will
effect large saving of power-plant fuel. The war
aroused all nations to a realization of the importance
of reliable and adequate industrial power, efficiently
produced, for maintaining industry and national
effectiveness at the maximum. The British Fuel
Research Board and the Nitrogen Products Committee
have made, and are continuing, comprehensive studies
of power development centralization. Our own Con-
gress has just provided $125,000 for investigation of a
possible super-power project for the Boston- Washing-
ton district.
There are installed now in the United States, or
nearing completion, central power stations aggregating
about 350,000-kw. capacity which use coal at or near
the mine mouth. These stations are laid out for an
ultimate capacity at least double that of the present
installation. They are consuming coal at an average
rate not far from 2.0 lbs. per kw.-hr. on the switchboard,
one-third less than the average consumption of public
utility power plants throughout the country, as shown
by statistical reports of the U. S. Geological Survey.
The advantages gained from the saving of freight
on coal and in reliability of service, add their weight
to those resulting from increased fuel economy, as
shown by the above figures. Reduction of overhead
and labor cost, and of the capital charges per unit
of power output unquestionably follows centralization
into large operating units, the gain being emphasized
5<=
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
by so choosing conditions as to permit of operation
under a high load factor. For this reason a super-
power station project may well take into account the
disposal of its output in part to chemical and electro-
chemical industries which can use power at night
or during the "off-peak" periods.
By-product recovery in connection with centralized
power development commends itself, on grounds of
conservation, to most careful investigation. Direct
coal-fired steam-turbo-electric stations afford a high
degree of fuel economy, but they waste entirely a
valuable national resource in the nitrogen of the coal,
vital to agriculture and to munitions of war. The
Nitrogen Products Committee of the British Ministry
of Munitions, after thorough investigation of various
systems of power production from coal, came to the
conclusion that the net cost of power in processes
involving carbonization or gasification of coal and
burning of the resulting coke and gas under boilers
was higher than in direct coal-fired steam turbine
stations, allowing a fair market value to the by-
products. Both high- and low-temperature carboniza-
tion were considered. The possibility of using gas
engines for power was dismissed by the Committee
as entirely impracticable for stations of the size neces-
sary for competitive operation under British conditions.
The reason advanced was the very high capital cost
of such installations and the cost for labor and repairs.
These conclusions do not necessarily apply to the
American problem of centralizing power development
for miscellaneous demand under a more widely varying
load. Gas engine power plants of 50,000-kw. capacity
on blast-furnace gas, in units of 2000 to 5000 kw.,
are operating successfully in this country at costs for
labor and repairs not materially higher than those for
equivalent steam turbine plants. It appears that
with due consideration of the returns from sale of by-
products and with due care so to restrict the scale of
operation as not to overload the by-product market,
a combination may be found practicable wherein gas
power would be used to meet the steady portion of
the plant load and coal-and-gas fired boilers to meet
the variable load. Surplus gas may be sold to the
gas companies for mixing with their own manufactured
outputs, or for reinforcing the waning supply of natural
gas.
The problem of choosing the best system for pro-
duction of gas and by-products in such central stations
is a many-sided one. To go into a detailed considera-
tion of it here would take us too far afield. A very
important phase requiring investigation is the mechani-
cal problem of proper design of engine to use gases of
high hydrogen content. This may or may not have
been sufficiently worked out at the present time. .
The gas-making process to be used in such an in-
stallation would be one permitting economical recov-
ery and high yield of ammonia, and at the same time
affording the highest thermal return from the coal.
Certain processes for the complete gasification of coal
by alternate production, in the same generator, of
distillation gases and of water gas by superheated
steam, have been developed to some extent and
show indications of being capable of higher thermal
efficiency than the two-stage gasification processes
now prevailing in coal-gas and water-gas manufacture.
Such a mixed gas would have a heating value of about
320 to 350 B. t. u. per cu. ft., a ton of coal yielding
about 50,000 cu. ft. if completely gasified. Ammonia
would be obtained in higher yield per ton than from
present carbonization processes. Other valuable by-
products would be recovered. The possible use of
oxygen produced electrolytically from off-peak power
on the plant to enrich the blast in such gas generators
is worthy of investigation for the sake of lowering the
content of nitrogen and hydrogen in the gas.
It may be found practicable in the future also, when
low-temperature carbonizing processes have been
further developed, to make use of them in such a cen-
tral station to a limited extent, possibly for raising
the heating value of the mixed gas and for producing
a clean, smokeless, solid fuel for disposal to the do-
mestic and small steam trade. Central power sta-
tions, distributing electric power only, are not likely
to displace steam plants for heating purposes, or for
chemical manufacture, dyeing, bleaching, etc. It
is desirable, however, in the interests of conservation
that carbonized fuels and gas be increasingly used for
this purpose.
gas manufacture — The trend in public gas supply
is toward the abolishing of lighting standards and the
substitution therefor of a thermal requirement lower
than has prevailed in the past. New Jersey has
recently adopted a 525 B. t. u. standard; the city of
Philadelphia has just agreed to a 530 standard; Massa-
chusetts has 528, and many other sections of the
country, including Chicago, are similarly progressive.
This means a lowering of the previous requirements
by 75 or 100 B. t. u., and will result in immense sav-
ings of oil in water-gas manufacture. It will permit
also the use of by-product coke-oven gas unenriched,
and in coal-gas manufacture the steaming of retorts
to give greater yields of both gas and by-products,
the increased gas yield permitting still more conser-
vation of oil in water gas. The cracking of oil in
water-gas manufacture is a wasteful process at best,
yielding soot and tar in place of available heat units,
and having lower thermal efficiency than the direct
burning of oil as fuel.
If gas companies were to be permitted still further
reduction of heating value, together with suitable
adjustment of rates to accord with the lower costs
of manufacture, there would undoubtedly result an
extension of the use of gas, particularly in the indus-
tries, with its attendant economies mentioned earlier
in this paper.
By-product coke ovens are steadily increasing in
number, but nearly half of the coke is still being made
by the old nonrecovery process, which burns, in
effecting the coking operation, 10 per cent of the coal
and all of the gas and by-products. If the 24,000,000
tons of coke now made annually in beehive ovens
were to be made in modern recovery ovens, it is safe
to say that a reduction of 8,000,000 to 10.000,000
tons in coal consumption would result, this being an
Jan., 19;
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
aggregate of the fuel equivalent of gas and tar saved,
increased coke yield, and improvement in blast-fur-
nace fuel efficiency. Ammonia and benzene recovery
would be an additional gain.
The conservation of coal by means of coking will
grow as the outlet for coke and by-products grows.
Extension in this field is not to be considered as limited
by the metallurgical demand for coke. Coke and
coke-oven gas as fuels, however, are likely to meet
strong competition eventually from cheap power
developed in central stations and from lower-cost
gas made by complete gasification processes.
colloidal fuel — Colloidal fuel deserves mention
in connection with fuel conservation. Colloidal sus-
pensions of pulverized coal in oil permit of the same
economies in application as either oil or powdered
coal alone, and have some advantages, notably per-
mitting the use of higher ash coals, higher sulfur oils,
and many carbonaceous waste products, concentra-
tion of heating value in relation to bulk, and decreasing
of fire hazard as compared to oil. It is of important
bearing, however, on the probable future development
of this new fuel to consider the oil reserves available
to the United States for fuel purposes.
SUMMARY
In general, why is fuel conservation to be needed
when our transportation systems shall become equipped
to deliver what is required? In the first place, effi-
ciency in the use of raw materials makes for increased
financial returns; secondly, waste promotes extrava-
gance and raises the cost of living; and lastly, our
high-grade fuel reserves are being exhausted at an
alarming rate. George H. Ashley, State Geologist
of Pennsylvania, estimates1 that practically all of the
easily workable coal beds of Pennsylvania, 6 ft. or more
in thickness, will disappear in 75 to 80 yrs. at the
present rate of increase in exhaustion. Low sulfur
coals for metallurgical purposes are becoming scarce,
so much so that steel men are investigating measures
for getting along without them. Yet the low sulfur
Pocahontas and New River coals are still sold in large
part for steaming purposes, where such low sulfur
content is not an essential quality.
There is a progressive tendency, however, in America
towards greater fuel economy, and future develop-
ments are likely to decrease materially our per capita
consumption.
DISCUSSION
Dr. Porter: It will perhaps bear repetition for the sake of
emphasis, that statistics show we are progressing remarkably
well in economic utilization of coal, and this paper accordingly
is not to be taken as a criticism of progress or lack of progress.
The consumption of coal per capita in the country has not in-
creased in the last few years, in spite of the fact that our iron and
steel production has gone up 50 per cent in 10 yrs., and
industrialization in general has very greatly expanded — the
production of automobiles, for instance, has multiplied itself
nearly ten times; also the standard of living to-day is much higher
in all classes than it was 10 yrs. ago, and yet the consumption
of coal per capita has remained practically on a level. Un-
doubtedly, therefore, we have made very material progress in
the efficiency of our application of coal.
1 By private communication supplementing published reports.
Dr. T. E. Layng: Mr. Chairman. I would like to ask Dr.
Porter about that 7. 1 per cent of coal used for gas making, export,
and bunkering. The exporting of coal has been severely criti-
cized; a great many people think it ought to be used in this
country. I should like to know about what percentage of that
7.1 per cent is exported.
Dr. Porter: My recollection of the figure for export this
year is that it is running now over 2,000,000 tons per month,
from tidewater, and a little less exported to Canada, which will
at that rate bring the total for this year close to 40,000,000 or
45,000,000 tons. The figures in the paper are for 19 17. The
export figures this year are very much higher than in 1917.
The export in 19 17, as I remember, was about 23,000,000 tons,
or 4.3 per cent of the total coal. Gas making required only
about 5,000,000 tons, or 1 per cent, and bunkering the balance.
GASOLINE LOSSES DUE TO INCOMPLETE COMBUSTION
IN MOTOR VEHICLES'
By A. C. Fieldner, A. A. Straub and G. W. Jones
Pittsburgh Experiment Station, U. S. Bureau op Min-e..
1 1 rrsBi
Pa.
The rapidly increasing use of motor vehicles in the
United States has introduced an entirely new problem
in the proper ventilation of tunnels, subways, and
other confined spaces through which such machines
must pass. This problem was brought to the atten-
tion of the Bureau of Mines last November by the
New York and New Jersey State Bridge and Tunnel
Commissions with reference to the ventilation of the
proposed vehicular tunnel under the Hudson River.
This tunnel, consisting of twin tubes 29 ft. in diameter
and 8500 ft. long between entrance and exit (Fig. 1),
presented an unprecedented problem in ventilation
both on account of its length and on account of the
traffic density, which is expected to reach a maximum
of 1900 vehicles per hour.
An exhaustive study by the tunnel engineers of all
available data on the amount and composition of
automobile exhaust gas disclosed very little informa-
tion on the percentage of carbon monoxide in motor
exhaust gas from the average run of automobiles and
trucks under actual operating conditions on the road.
It was well known that carburetor adjustment and
other operating factors changed the percentage of the
poisonous constituent, carbon monoxide, from prac-
tically o to 1 2 or 13 per cent; but no safe estimate could
be made of the most probable figure without further
investigation.
A series of tests was therefore undertaken in which
passenger cars and trucks were tested in exactly the
same condition as furnished by the owners from whom
they were borrowed. No change was made in car-
buretor adjustment or any other operating condition,
the prime object being to obtain information on existing
operating conditions and not the ideal conditions of
careful adjustment under which the usual test of the
automotive engineer is made. For this reason the
data are of especial value in showing the proportion
of gasoline wasted by the average automobile owner
and truck operator through imperfect combustion.
1 Published with the permission of the Director, U. S. Bureau of Mines
and of the Chief Engineer of the New York and New Jersey State Bridg
and Tunnel Commissions.
5-
TEE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No.
+/£0S£r C/TY
HUDSON RIVER VEHICULAR TUNNEL
DIAGRAM SHOWING METHOD OF VENTILATION
PROFILE 4 SECTION
SECTION Or ONE TUNNEL
Fig. 1 — Plan, Profile,
PLATE N5
1 Sections op thk Hudson River Vehicular Tunnels
»&*s
METHOD OF CONDUCTING TESTS
All cars were tested in the same condition as re-
ceived, and with the same .brand of gasoline that the
car was using. Fig. 2 shows a 2.5-ton truck equipped
with gasoline measuring apparatus (in front of driver's
seat) and exhaust gas sampling tube (back of cab).
GASOLINE MEASURING APPARATUS The gasoline
measuring apparatus shown in Fig. 3 was connected
directly to the carburetor and to a reserve supply of
gasoline, v, through the copper pipes n and c, respec-
tively.
As the car crossed the boundary lines of the test
course at the predetermined speed for the test, the
gasoline feed was switched from the reserve supply
to the measuring tube /, by closing the cock e and
opening q. At the end of the test course, a reverse
operation of these cocks switched the supply hack to
the reserve supply tank.
The exhaust gas pressure was sufficient to maintain
a rapid stream of gas through the heavy-walled rubber
tube b connected to the glass tee a on the sampler
board. The main stream of exhaust gases passed on
through the rubber tube b and was discharged into
the atmosphere through the water seal c, thus pre-
venting any air from being sucked back into the
sample.
The exhaust gas sample was collected continuously
at a uniform rate over the whole period of the test,
in a 250-cc. glass sampling tube connected to the down-
ward branch of the tee a. One observer gave his
entire attention to regulating the flow of the water
from the sample tube, by adjusting the screw clamp
at the lower end of the tube. A 5 per cent solution
of sodium chloride previously saturated with exhaust
gas was used.
jkwrmmu. .
I TEST CAft 1 .
flfeS
saH*
Riiniii
MiUli
it
Bjfl 1
•J~tm
^}
t ~~ ■ - —
^^^B
Fig. 2 — 2.5 Ton Truck, Loaded and Equipped for Road Tests
SAMPLING AND ANALYSIS OF EXHAUST GASES The
exhaust gas sampling apparatus is shown in Fig. 4.
A 0.25-in. copper tube, g, bent at right angles, with
the opening turned toward the engine, was introduced
into the exhaust pipe between the engine and muffler.
Fig. 3 — Gasoline Measuring Appar
The samples were analyzed in duplicate for COj,
02, CO, H2, N2, and CH4 on a laboratory type Burrell-
Orsat apparatus1 as used in the Bureau of Mines for
Burrell and F. M. Seibert, "The Sampling
es and Natural Gas," Bulletin 42 (1913), 43.
ad Examination.
Jan., 1921 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
53
) Sp. Gr. Baui
0.713 66.
0.731 61.
0.730 61.
0.796 45.
Benzene mixture.
ate Analyse
Hydro-
n gen
15.7
15.7
14.8
117
First
Drop
Table I — Analyses of Gasoline Used
Distillation in 100 Cc. Engler Flask, T<
30%
176
214
214
214
40%
201
239
237
228
50%
225
266
259
248
60%
250
293
282
271
347
327
345
90%
381
394
363
381
239
282
259
264
5.0
3.0
3.0
2.0
complete gas analysis. The carbon dioxide was
absorbed in potassium hydroxide solution; the oxygen
in potassium pyrogallate; the carbon monoxide in two
bubbling pipets in series, containing acid cuprous
chloride solution; and the hydrogen, methane, and
any residual carbon monoxide were determined by
slow combustion in the presence of a hot platinum wire.
Fig. 4 — Exha
MPLING ApPA
In this method of analysis any gasoline vapor and
other hydrocarbons appear as methane. In other
words, the analysis gives the equivalent methane
value for all the hydrocarbons in the exhaust gas,
and the result is correct as regards carbon content for
computing the total volume of exhaust gases from the
gasoline consumption and the carbon content of the
gasoline. This relation was checked to within 6 per
cent by actual measurement of exhaust gas in a 50
cu. ft. container.
The determination of gasoline vapor as methane
causes the hydrogen value in the analysis to be some-
what less than its true value. This error in the hydro-
gen value has no effect on the calculation of the true
value of CO, CO2, and CH4 equivalent of total hydro-
carbons.
gasoline used — -Each car was tested with the same
brand of gasoline as the driver was using when the
car was submitted for test. Analyses of these various
brands are given in Table I.
test conditions — Tests were made under the
various conditions which might prevail in the tunnel,
at different times, as for example:
Car at rest with engine idling.
Car at rest with engine racing.
Car accelerating from rest to 15 mi. per hour on level and up a 3 per
cent grade.
Car running 3 mi. per hour on level grade, up 3 per cent grade down
3 per cent grade.
Car running 10 mi. per hour on level grade, up 3 per cent grade, down
3 per cent grade.
Car running 15 mi. per hour on level grade, up 3 per cent grade down
3 per cent grade.
The level and 3 per cent grade courses were each one
mile long; the surface was asphalt on the grade course,
and part asphalt and part macadam on the level
course.
Trucks and 7-passenger cars were tested with both
light load and full load, the light load consisting of
two observers, driver, and the necessary apparatus.
One hundred trucks and passenger cars were tested
in the entire investigation; twenty-three were tested
under winter conditions, and seventy-seven were tested
under spring and summer conditions.
RESULT OF TESTS UNDER WINTER CONDITIONS
A summary of the results of tests of twenty-three
passenger cars and trucks under winter conditions is
given in Tables II, III, and IV.
Table II — Average Results of Tests on Eleven 5-Passenger Cars
Condition
of Test
Engine racing
Engine idling
Up 3 per cent
grade :
15 mi. per hr.
10 mi. per hr.
3 mi. per hr.
Down 3 per cent
Com- Lbs.
plete- Air
Mi. ness of per Lb.
per Com- Gaso-
Gal. bustion line
15 :
Level i
15 r
. per hr.
. per hr.
. per hr.
-ade:
i. per hr.
. per hr.
. per hr.
24.5
22.8
9.9
16.9
16.9
7.5
12.6
13.0
12.2
12.3
12.3
12.9
13.4
12.7
12.6
Analysis of Exhaust Gas
. Per cent by Volume
CO2 Oi CO CHi H2
9 1 1.5 6.9 0.8 3.0 ;
10.2
9.9
9.8
9.5
8.6
9.5
9.3
9.3
9.1
1.4
1.4 .
1.5 6.0
2.2 5.6
1.9 6.3
1.6 6.7
0.6
0.5
0.6
6.5 0.9
7.0 0.7
" 0.7
2.6
2.6
3.0
0.8
0.6
0.6
2^7
,i'ii
78.8
79.2
79.6
79.3
78.8
79.0
Table III — Average Results of Tests on Seven 7-Passenger Ca
Condition
of Test
Engine racing
Engine idling
Up 3 per cent
15 mi. per hr.
10 mi. per hr
3 mi. per hr.
Down 3 per cent
15 mi. per hr
10 mi. per hr.
3 mi. per hr.
Level grade:
15 mi. per hr.
10 mi. per hr.
3 mi. per hr.
Com- Lbs
plete- Air
Mi. ness of per Lb
per Com- Gaso-
Gal. bustion line
Analysis of Exhaust Gas
* Per cent by Volume —
CO; O- CO CH4 Hi
7.3 3.5 7.8 1.4 2.9
8.0 4.3 6.3 1.2 2.0
16.9
19.4
9.4
14.0
14.9
15.3
6.4 6.0 6.8
6.9 5.0 6.:
6.9 5.0 6.3
2.4
2.2
2.4
6.5 0.9 2.8
6.4 1.1 2.8
7.0 1.0 3.0
54
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
^ ft
< °<
*'
" s. -,' % J- 3!v "~ ^4*
?
as%
- =. S S ^ *< 2* r'*^ =5P--dL
^ "^
ip-^ 5^ -t:
Vj 5
I — 1 — ! — > — ! — ! — J-—1 — ! — ! — ' ' '^ ' — ' — ' — ' — ' ' J — ' — ' — \_
erageso posse nper roaring cars
i s=~ — ^ ° ° Averages of II- 5 passenger touring cars
§
7 ^ X X Averages of 5lrucks; l-lton,h$ ton, _
T_ \ . Z speed trucks and 1- 10 passenger bus.
\
C5
~1 \ ~[_ S,
-h-"
\
/ .. v / *s
5
V
' ^'"v X _r V
*
\
A .-^__^ i, V 7- -.r- V.
\
7 T '^%, \ -f 2"v \ 4-
f
t * % s NZ v>^^ \,_
- Z N^ ^2' ^sO^
£ -
C-
^-/ ^ J' 'C'Ov,
s
v
tt ^S ;VSJ'
^
'
/
/
it.
Tr ir ^^ >r_
■3
\
S
iic5-'
>
y
10 ,nn
$
L S, 3.--<l ""^
?
L J>=.«^'"!* ^ ;>t, ^S.
V
, ^< "H^;'~. ^^^ >' *s j— 2E~i
^50'
!s»
S
7 ? J<s^^-,r-^r:.ir--s-^ ~1-ii
<^
u^
.<$*-- ^ — -^,
1
0
S.
0
5
>
D
%
a
m
j
PE
r
rcl
•^ S t^ t* 1^ 8 '^ t* "H
»1 ^ !Q ^ *>> ^ ^ <•>
VI GRADE. UP 3 PER CENT GRADE ON LEVEL GRADE
(DHR Winter
Fig. 5 is a graphical presentation of the important
figures as regards tunnel ventilation, namely, the aver-
age per cent of carbon monoxide in the exhaust gas,
the gallons of gasoline consumed per hour, and the
cubic feet of carbon monoxide per hour.
Table TV — Average Results of Tests on Five Light Trucks
Condition
per
of Com-
of Test
Gal.
bustion
Engine racing
64
Engine idling
57
I p 3 per cent
grade:
15 mi. per hr.
11.6
73
10 mi. per hr.
10.7
64
3 mi. per hr.
5.9
63
Down 3 per cent
grade:
15 mi. per hr.
21.6
63
in mi. per hr.
17.1
56
3 mi. per hr.
7.7
56
Level grade:
15 mi. per hr.
is.:
67
10 mi. per hr.
12.9
63
3 mi. per hr.
6.1
62
er Lb.
Gaso-
line
CO*
Analysis of Exhaust Gas
— Per cent bv Volume .
Ot CO CHi Hi N:
11.3
12.0
8.3
6.6
2.0
4.2
7.7
7.1
1.2
2.1
4.0
3.7
76.8
76.3
12.5
11. 0
11.2
9.6
9.0
8.1
1.5
1.3
1.6
6.2
7.0
8.5
0.6
1.3
1 .2
3.0
4. 1
4.4
79.1
76.2
12. 1
ii .:
12.3
7.5
6.5
6.5
3.1
4.1
3.6
7.1
7.7
7.5
1.4
3^6
3.4
77.4
7(. 1
76.8
11.8
12.0
12.0
9.0
7.7
7.4
1.5
2.1
2.9
7.0
8.0
7.7
1.1
1 .3
1.3
3.4
3.8
4.1
78.0
77.1
76.6
discussion of results of tests — It will be noted
from the plotted results that the average percentage
of carbon monoxide for each class of vehicles varies
between 5 per cent as a minimum and 9 per cent as a
maximum, the larger percentages tending to be pro-
duced when the engine is racing, idling, or running on
light load on the low gear at 3 mi. per hr. However,
the greatest amount of carbon monoxide per hour is
generated under conditions of greatest load, i. e.,
when accelerating or running up grade at the highest
speed.
The relative quantity of carbon monoxide produced
depends primarily on the gasoline consumption as
shown at a glance by the similar rise and fall of the
"gasoline" and "cubic feet of carbon monoxide"
curves.
The average percentage of carbon monoxide under
all conditions of test for each class of vehicles was
5-passenger cars 6.3; 7-passenger cars 6.8; and light
trucks 6.9.
These values are consistently higher than reported
by previous investigators. The most extensive road
tests heretofore made in this country are those re-
ported by Herbert Chase* in 1914- A comparison
of his results with the Bureau of Mines tests is given
in Table V.
Table V — Comparison
Hxhaust Gas Analyses of Tests by
y the Bureau op Mines
Average Exhaust Gas Analyses
-Per cent by Volume-
r — Carbon Monoxide—*
Chase B. of M. Din*.
Cars standing, engine idling 2.6 7.1 4.5
Cars accelerating to 10 mi.1
per hr. from rest 1.9 5.6 3.7
Car^ running 10 mi. per hr.
on level grade 2.3 6.7 4.4
Cars running 15 mi. per hr.
on level grade 2.5 6.3 3.8
Average 2.3 6.4 4.1
1 15 mi. per hour in Bureau of Mi
-Carbon Dioxide
Chase B. of M. Diff
8.4
10.1
9.7
9.5
9.4
9.5
8.8
9.0
0.3
0.6
0.9
0.5
0.6
'Exhaust Gas Analys
tests.
for Economy," The Automobile, 30 (Februa
Jan., iqj i
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
55
The average of all comparable tests shows 0.6 per cent
more carbon dioxide and 4.1 per cent less carbon
monoxide in the Chase tests than in the Bureau of
Mines tests.
The cause for the large difference in carbon monoxide
percentages is not clear, in view of the agreement in
the carbon dioxide results. If the carburation and
combustion of the less volatile present-day gasoline
is less efficient than in 1014 we should expect a corre-
sponding difference in the carbon dioxide percent-
ages.
Hood, Kudlich and Burrell,1 have shown that the
proportion of carbon monoxide in exhaust gases varies
from o to about 14 per cent, the amount depending
on a number of variables, chief of which are:
( 1 ) Ratio of air to gasoline
(2) Completeness of vaporization and mixing
(3) Speed of engines
(4) Temperature of air and jacket water
(5) Quality and time of spark
(6) Degree of compression
(7) Quality of gasoline or motor fuel
In view of this large number of variables it is not
surprising that extremely large variations in exhaust
gas composition were obtained in testing motor vehicles
taken from ordinary service without any adjustment
prior to test and driven in a variable manner with foot
accelerator or hand throttle by different drivers over
an approximately smooth course, but yet one with
some rough places requiring opening and closing the
throttle to maintain a constant speed.
It is, therefore, not possible to draw conclusions on
the effect on exhaust gas composition of the various
factors just enumerated, except with regard to the
first one, namely, "ratio of air to gasoline," or carburetor
adjustment.
EFFECT OF CARBURETOR ADJUSTMENT A Study of
all the tests shows that the variation in exhaust gas
composition due to carburetor adjustment is far greater
than any other factor; they do not throw much light
on the advantage of any particular make or type of
carburetor, nor should any conclusions be drawn
as to the merits or demerits of any particular make
of car.
Table VI — Best and Poorest Results Obta
SENTATIVE MAKES OF PASSENGER C
(All cars loaded)
&m -a § fl
1* ° :« si
3 a S. £« til
■gj g "~ So
to S P< fc
15 27.30 105.8 100
15 13.26 84
15 18.61 66.8 93
IS 11.16 61
15 15.39 44.5 90
15 10.66 59
10 6.55 36.2 87
10 4.81 65
15 10.26 49
O § fr<
1 C 5-passenger
9 C 5-passenger
11 G 7-passenger
10 G 7-passenger
84 X V.-t. truck
76 X »A-t. truck
38 Y 3.5-t. truck
57 Y 3.5-t. truck
44 D 5-passenger
Exhaust Gas Analysis
^J
, — Per cent by Volume — .
CO- O2 CO CH. H2
<
13.0 2.6 0 0 0
16.7
11.8 0.8 3.7 0.3 1.6
13.5
9.3 5.4 1.3 0 0.1
20.1
7.5 2.1 9.3 1.4 4.0
10.7
10.7 3.9 1.7 0.5 0.2
16.6
7.1 0.7 10.7 1.0 5.1
10.3
12.9 0.3 1 .9 0.8 0.4
13.9
7.5 0.8 10.6 1.0 4.9
10.2
5.3 1.0 13.2 1.9 7.1
9.0
Table VI gives a comparison of the best and poorest
tests obtained on several well-known makes of passenger
and greatest mileage of any car tested. Car No. 44,
cars and trucks. Car No. i had the best gas analysis,
also a 5-passenger car, had the poorest gas analysis
and the lowest mileage in its class. Both cars operated
without any apparent difficulty throughout the tests.
Car No. n did not operate smoothly and lacked flexi-
bility at low speed due to the mixture being too lean.
However, the mileage per gallon of gasoline was much
higher than the other cars in the same class. At
speeds above 15 mi. per hr. it operated smoothly and
gave a good illustration of the tremendous quantity
of fuel that may be saved by using lean mixtures.
It should be noted that in each case the car with the
leaner mixture shows the largest mileage per gallon of
gasoline. The percentage increase in mileage ranges
from 36 to 106 per cent.
The effect of various carburetor adjustments on an
individual car is shown in Table VII.
Table VII — Effect of Carburetor Adjustment on Gasoline Con-
sumption and Exhaust Gas Analysis
4-cylinder roadster, engine 41/b in. bore X 41/? in. stroke; Johnson
carburetor; intake air and manifold heated; using gasoline 66.4° Baume,
distillation 10%, 127° F.; 50%, 225° F., dry, 441° F.; average 239° F.
Tests at 15 mi. per hr. ascending a 3 per cent grade of asphalt in good con-
dition.
Gal.
per
Miles
per
Gal.
14.9
13.9
10.6
Mile
0.067
0.072
0.094
0.1142
-Exhaust clear, mixture too Ie
Qxhaust Gas
Analyses, Per cent
CO. O2
13.4 1.7
12.0 1.4
10.2 0.3
6.5 1.2
CO CH.
0.2 0.0 83.5
2.0 1.1 0.0 83.5
6.4 0.8 2.4 79.9
1.6 1.0 6.4 73.3
ithout
9.9 56
to operate
choke 1/4
■ .,1 .
"Gasoline Mine Locomotives
u of Mines, Bulletin 74 (1915)
Relation to Safety and Health,'
b — Exhaust clear, operation satisfactory,
part of test.
c — Exhaust slightly smoky; operation satisfactory. Car had good
"pick-up."
d — Smoky exhaust; mixture seemed too rich for satisfaT*tory operation.
Before putting this car. through the standard series
of "road tests the driver, an automobile mechanic, was
asked to place the carburetor in good adjustment.
He set it after the engine was warmed up to running
conditions, at i7/i6 turns of the needle valve. As
shown in the table this setting produced 6.4 per cent
carbon monoxide and 10.2 per cent carbon dioxide, a
little better than the average analysis of all the cars
tested. Tests were then repeated under identical
conditions with both richer and leaner settings. It
was found that i1/* turns of the carburetor needle
gave 12 per cent C02 and 2.0 per cent CO; and 31 per
cent greater mileage; also the car operated satisfac-
torily.
This test is typical of the great majority of the
passenger cars and trucks tested, they were invariably
adjusted safely on the rich side for greatest flexibility
of operation rather than for maximum economy of
gasoline.
REASONS FOR EXISTING USE OF RICH MIXTURES
One pound of ordinary motor gasoline of to-day,
such as was used in the tests just described, requires
approximately 15 lbs. of air for complete combustion.
56
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
•=t^
3§
■\
\ *'-
/
s "
;
/
\«!
/
<
/
/
-k
r
,5
N
i
>s
&
^"^
p
1
|
\ 1
/
\ \
/
\ '
,
/
/
/
1
\
/
,
1
J
!
zr.
18 n lb 15 14 13 12 II 10 9 B
RATIO OF AIR TO GASOl/Nf, POUNDS
ig. 6 — Curves Showing Relation between Braki
Thermal Efficiency at Various Air-Gasolin:
Berry
Horse Power and
Ratios. After
The maximum thermal efficiency is obtained at about
16 lbs.1 of air to 1 lb. of gasoline, and the maximum
power with 12 to 13 lbs. of air.2 Herein lies the reason
for the use of rich mixtures. The average driver
demands first of all power and flexibility of operation.
He sets his carburetor adjustment rich enough to
give good operation with a cold engine and for slow
driving in heavy traffic, with plenty of reserve power
for hill climbing and bad road. If he errs somewhat
on the rich side it does not become manifest in loss of
power, but only in the increased gasoline consumption,
which in many instances does not concern him at all.
An inspection of the average thermal efficiency and
power curves of Fig. 6 shows that the proportion of air
in the mixture can be reduced to 9.0 lbs. of air to 1 lb.
of gasoline with a loss of only 9 per cent in power,
although economy and efficiency are tremendously
reduced.
Fig. 7 shows the relation between the air-gasoline
rates and the percentage of carbon monoxide in the
exhaust gas for the first 23 passenger cars and trucks
tested at 15 mi. per hr. running up a 3 per cent grade.
The air ratios varied from 15.8 with about 1.0 per
cent carbon monoxide, to 9.7 lbs. air with 12.3 per
cent carbon monoxide. The average air-gasoline
ratio was 12.4, with an average carbon monoxide per
cent of 6.3, practically the exact figure for maximum
power. Obviously, carburetors are adjusted in prac-
tice for maximum power and not for maximum thermal
efficiency and economy of gasoline.
The average loss of gasoline due to the continuous
operation of a car at the point of maximum power is
shown in the accompanying computations from average
exhaust gas analyses, heat in the gasoline, and heat
in the unburned exhaust gas constituents.
1 With this mixture the engine develops about 85 per cent of its max-
imum power.
2 O. C. Berry, "Mixture Requirements of Automobile Engines," J,
Soc. Automotive Euc.. 5 (1919), 364.
1 nl ..hi dioxide
Level Grade
Per cent
8.9
2.3
Ascending 3 Per
cent Grade
Per cent
9.6
1 .3
0.9
0.6
Total
Cu. ft. exhaust gases at
29.92 in. Hg
65'
I-
100.0
ind
. . . 988
100. 0
Composition of Gasc
Sp. Gr 0.713
Carbon 84.3 per cent
Hydrogen 15.7 per cent
Calorific value . ... 21.300 B. t. u. per lb.
130,000 B. t. u per gallon
Exhaust Ga
prom I Gal. Gasoline oi
988 X 6,3 = 62.2 i
988 X 0.9 = 9.1 i
988 X 3.0 = 2.9 .
Level Grade Tests Contains
a. ft. CO
j. ft. CHi
J. ft. Hi
Total Heat in Unburned Gases pe
B. t. u.
62.2 X 320' = 19,900
38,500
1 Gross B. t. u. per cu. ft. at 65° F. and 29.92
38,500
130,000
29.6 per cent of the total heat of the gasolii
the form of combustible gases.
Gallon Gassune
= 29.6 per cent
goes out ia the exhaust
o
>- h
5 h
<:>
o
',.
i
0
0
<
ox\
8
*
1
1 _
\
c
o
5<
N^C
X
>
t
0
«.
X
pounds of air per pound of gasoline
Fig. 7 — Curve Showing Relation between Air-Gasoline Ratio and
Carbon Monoxide in Exhaust Gas of 23 Cars Tested at 15 Mms
per Hour Running up a 3 Per cent Grade
RESULTS OF TESTS UNDER SPRING AND SUMMER
CONDITIONS
While the data just given for winter conditions
show surprisingly large losses due to incomplete com
bustion, incomplete returns on the summer tests show
even larger losses. As shown in Table VIII, passenger
cars and the lighter trucks average from 6.0 per cent
to 7.6 per cent carbon monoxide.
Table VIII — Comparison of Percentage of Carbon Monoxide in
Exhaust Gas in Winter and .summer
Average Per cent Carbon
Type of Car . — Monoxide in Exhaust Gas1 — .
Winter Summer
5 -passenger car 6.3 7.6
7-passenger car 6.8 7.4
Trucks up to 1.5 tons 6.9 7.7
Trucks 1 .5 to 3 tons ■. 6.9
Trucks 3.5 to 4.5 tons 6,3
Trucks 5 tons and over 6.0
' Average of all conditions of test previously described.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
5 7
It appears that most cars are adjusted to start easily
in cold weather and then are permitted to remain the
same during the entire summer, thus increasing the
wastage of gasoline during the period of greatest con-
sumption.
Probably 50 to 75 per cent of the present daily loss
of gasoline due to the prevalent use of rich mixtures
could be prevented by proper adjustment of existing
forms of carburetors. Unfortunately, most drivers
do not care to change even a simple manually controlled
adjustment from the dash. They set it rich enough
for the heaviest load and then leave it the same for
all duties.
AUTOMATIC CARBURETOR NECESSARY
It is hoped that the results of these 23 tests and the
remaining 78 which will be published at an early date
will serve as a stimulus to automotive engineers to
design an automatic carburetor as suggested by
W. E. Lay,1 who states:
The ideal carburetor would be arranged so as to
supply primarily the mixture giving the best efficiency and
automatically supply the necessary additional fuel only when
operating conditions require it. The provisions made should
be so adequate that the economy under proper operating condi-
tions will never be sacrificed to obtain more power or better
operation under exceptional conditions.
SUMMARY
Road tests under winter conditions for the purpose
of determining the amount and composition of motor
exhaust gas from automobiles and trucks of various
sizes when operated on grades and at speeds similar
to those that will prevail in vehicular tunnels have
shown that:
(1) The exhaust gas composition of individual
machines varies greatly, and the controlling factor
is the air-gasoline ratio produced by the carburetor
adjustment.
(2) The percentage of carbon monoxide for the
majority of cars lies between 5 and 9 per cent.
(3) The average percentage of carbon monoxide
for 23 cars tested was 6.7 per cent, which is practically
the ratio for developing maximum power.
(4) The combustible gas in the average automobile
exhaust from one gallon of gasoline amounts to 30
per cent of the total heat in a gallon of gasoline.
(5) The great majority of motor cars and trucks
are operated on rich mixtures suitable for maximum
power but very wasteful from the standpoint of gaso-
line economy.
(6) On the average, carburetors are set in the winter
and not changed in the summer, as shown by the
higher percentages of carbon monoxide found in the
summer test.
(7) A simple and convenient dash adjustment for
instantly throwing a carburetor adjustment from the
condition of maximum thermal efficiency to maximum
power for steep hills and for starting the machine
would probably result in saving 20 to 30 per cent of
1 "Saving Fuel with the Carburetor," J. Soc. Automotive Eng., 7 (1920).
189.
the gasoline used, not a small item when we consider
the total gasoline used by the 7,500,000 automobiles
and trucks operating in 1919.
(8) An automatic self-changing carburetor which
gives rich mixtures for power only when needed would
be the solution of the problem of saving gasoline losses
from incomplete combustion.
DISCUSSION
George G. Brown: Mr. Chairman, I have been very much
interested in this proposition of combustion gas in the car-
buretor. Back in 1913, the time so many analyses were made,
the truck drivers were more careless with their carburetors than
they are now, although we found that some of them did fairly
well day after day under the same truck driver. One reason
for this change is that the carburetors have been improved.
But here are a few facts which may be interesting and which
have been checked by the Royal Automobile Club of England.
They have found the maximum power for a car runs about 12
parts by weight of air to 1 part of gasoline. That would give
an excess of gasoline, and therefore some carbon monoxide.
The maximum thermal efficiency runs about 17 parts of air by
weight to 1 part of gasoline. That is an excess of air; and for
complete combustion, depending on the kind of gasoline used,
it runs about 14.5 to 15 parts of air. As has been pointed
out, the key to the whole situation is really in the design of a
carburetor. A properly designed carburetor should give 12
parts of air to 1 part of gasoline when climbing a hill, and when
running on a level it should automatically give 17 parts of air
to 1 part of gasoline. In other words, what is wanted is the
uniform mixture for maximum economy; we want what most
carburetors do not give, a light mixture when the engine is
running light, when running at high speeds, and a heavy mix-
ture when the engine is running slow on heavy load. Most of
the carburetors on the market at the present time have just
the reverse action, because at a higher velocity all of the air
going through the carburetor causes a greater proportion of
gasoline to be drawn into the mixture than is the fact under
reverse conditions, so that in going at higher speeds we get a
richer mixture. At the point where you get the richest gas you
want the weakest.
We have been working on this, and we have got
thus far: We can get a light mixture when the engine is running
light and a heavy mixture when it is running heavy. If we can
get a carburetor on a car so that it will answer automatically
and scientifically all changes in road conditions and all changes
in temperature, and if we can then locate the carburetor so that
the driver cannot adjust it except with the aid of a service man,
I think we have gone a long way toward getting the maximum
efficiency out of the engine. We have got everything lined up
except the temperature, and we can work that out very
shortly.
I am not prepared to go into the theory of the whole proposi-
tion with you but, I thought I would bring this out at this time-
not only the adjustment of the carburetor, but what you want is
a scientific, fool-proof carburetor, and there is nothing of that
kind that I know of in the market at the present time.
Mr. R E. Wilson: I would like to ask if the amount of car-
bon monoxide is going to make the ventilation in that tunnel a
particularly difficult matter?
Mr. FiELDNER: No, it doesn't make it particularly difficult,
but it will take some power and machinery to do it. The engi-
neering difficulties are not so great as one might think. They
have to put through about 1,500,000 cu. ft. of air per minute.
In reference to Mr. Brown's remarks on carburetors, it is inter-
esting to point out that the average of the air-gasoline ratio on
the 10 cars tested by the Bureau of Mines was something like
5*
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13. No. 1
12.5; in other words, carburetors are adjusted for maximum
power rather than maximum thermal efficiency.
.Mk. Brown: We figured out a few years ago that running
on a theoretically perfect combustion basis, that is, about 15
parts of air to 1 of gasoline, the mileage of a Ford would be a
little over 26 mi. per gal.; if you are getting 20 mi. per
gal. on a Ford you are getting what should be obtained without
any excess gasoline, without any carbon monoxide in your
exhaust. We have obtained as high as 38 mi. per gal.
with careful adjustment and careful driving, but over a long
period of driving through streets, etc., we have averaged over
28 mi. per gal. On the basis of getting a very light mixture,
we can get 26 mi. to a gallon on a Ford. Usually a man makes
22 or 23. A man in Long Island told me the best he knew was
19.5. There is a tremendous saving to be made there, aside
from the fact that we are relieving the engine from pumping
through 1,000,000 cu. ft. of air in a minute, because we have
found an average of less than 1 per cent carbon monoxide under
all conditions.
ENRICHMENT OF ARTIFICIAL GAS WITH NATURAL GAS
By James B. Garner
Director op Research and Development Department, Hope and
Peoples National Gas Companies. Pittsburgh, Pa.
ABSTRACT
The project of enriching artificial gas with natural
gas is of widespread interest because of the possibility
it offers of providing a supply of a clean domestic fuel
gas, uniform in quality, and of sufficient volume to
meet the requirements of the public. This is par-
ticularly the case in regions where natural gas has been
used.
There are in nature three potential sources of raw
materials adequate for the production of a future do-
mestic supply of manufactured gas: bituminous shale,
oil, and coal. Artificial gas, as produced on a com-
mercial scale, consists of the following varieties: shale
gas, oil gas, producer gas, water gas, carbureted water
gas, coal, and coke-oven gas.
Shale gas has been made and utilized with some de-
gree of efficiency in Scotland, and considerable experi-
mental work has been done in the United States look-
ing toward the development and utilization of our
vast beds of bituminous shale. With our present lack
of engineering and technical knowledge regarding the
use of bituminous shale as the future source of an
adequate supply of manufactured gas, its geographic
location and availability is such that bituminous shale
cannot now be considered as an immediately available
raw material.
Oil gas is the domestic gas of San Francisco, Oak-
land, Los Angeles, Portland, Tacoma, and San Diego.
Oil is used as the basis of gas manufacture in these
western cities because of the nonavailability of cheap
coal, while cheap oil is available. In all other sec-
tions of the United States, gas-oil or other products
from petroleum are so expensive that the manufacture
of oil gas is economically prohibited.
Producer gas, water gas, carbureted water gas,
coal, and coke-oven gas have all been made and used
with greater or less success for many years past.
Coal seems to be the only raw material which is at
present available as a basis for a future gas supply.
Producer gas is unsuited for use as a domestic gas for
two reasons:
1 Its high content of inert nitrogen, and (2) the excessive
cost of cleaning, cooling, and distributing.
Coke-oven and coal gas of a high quality are made,
but on account of the cost of installation and non-
flexibility of the plants wherein these gases are pro-
duced, these processes of manufacture are unfitted for
use in meeting the peak-load requirements of an ade-
quate domestic supply.
Blue water gas, although lower in heating value than
coke-oven or coal gas, can be made most economically;
and in a plant which is cheap in its cost of installation
and flexible in its operation, blue water gas is at present
the only rational basis for an adequate supply of
clean, uniform fuel gas to meet peak-load public re-
quirements. Blue water gas carbureted by means of
gas oil cannot, under present market conditions of
crude petroleum, be the kind of commercial gas for
an adequate public supply. In addition, this use of
the waning supply of crude petroleum is far from the
conservation of one of our greatest natural resources.
In order to carburet water gas of an initial heating
value of 325 B. t. u. per cu. ft. so that it will have
a heating value of 570 B. t. u. per cu. ft., it is
necessary to use 3 gal. of gas oil per 1000 cu. ft. of
gas. The present market on gas oil is 12 cents per
gallon. The enriching of 1000 cu. ft. of gas thus costs
the producer 36 cents without any overhead, produc-
tion, or depreciation charges. Natural gas, as pro-
duced in the Appalachian and Mid-Continent fields,
has an average heating value of 1100 B. t. u. per cu.
ft. It can readily be seen that less than 80 cu. ft.
of natural gas has an enriching value equal to one
gallon of gas oil. Natural gas can be mixed with blue
water gas easily, safely, and without any overhead,
production, and depreciation charges, and is, therefore,
the ideal enricher of water gas, in regions where nat-
ural gas is available.
The manufacture of a domestic supply of water gas.
enriched with natural gas, serves two purposes:
(1) It conserves in the highest possible manner our natural
resources of coal, oil, and gas.
It insures to the public an adequate supply at all times "I
a clean, uniform gas at the lowest possible cost.
Natural gas companies should no longer sell natural
gas as such at ridiculously low rates, but should utilize
it in the highest possible way, viz., as a means of en-
riching artificial gas. Such use of this natural resource
will insure to the public, for many years to come, a
supply of gas at a cost otherwise impossible.
THE CHARCOAL METHOD OF GASOLINE RECOVERY
By G. A. Burrell, G. G. Oberfell and C. L. Voress
Inasmuch as this paper has already been published
in another journal it is not included among the sym-
posium papers here.
Jan.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
ORIGINAL PAPERS
NOTICE TO AUTHORS: All drawings should be made with
India ink, preferably on tracing cloth. If coordinate paper 'is
used, blue must be chosen, as all other colors blur on re-
duction. The larger squares, curves, etc., which will show in
the finished cut, are to be inked in.
Blue prints and photostats are not suitable for reproduction.
Lettering should be even, and large enough to reproduce
well when the drawing is reduced to the width of a single column
of THIS JOURNAL, or less frequently to double column width.
Authors are requested to follow the Society's spellings on
drawings, e. «., sulfur, per cent, gage, etc.
STUDIES ON THE NITROTOLUENES. V— BINARY
SYSTEMS OF o-NITROTOLUENE AND
ANOTHER NITROTOLUENE'
By James M. Bell, Edward B. Cordon, Fletcher H. Spry and
Woodford White
University of North Carolina, Chapel Hill, N. C.
Received November 8, 1920
The third paper of this series, by Bell and Herty,2
records the results of studies of the binary systems
of the components: ^-nitrotoluene (MNT), 1,2,4-di-
nitrotoluene (DNT), and 1,2,4,6-trinitrotoluene (TNT).
The present paper contains the results of work upon
three binary systems in each of which o-nitrotoluene
(ONT) is one of the components, and one of the above
nitrotoluenes is the other component.
PURIFICATION OF THE NITROTOLUENES
Crude MNT was crystallized several times from
hot alcohol solution, filtered by suction, and allowed
to dry in a warm place. A constant melting point
(51.3 ° corr.) accorded well with the earlier work.3 In
a similar way DNT and TNT gave constant melting
points of 60.55° (corr.) and 80.35° (corr.), respectively.
Crude ONT was distilled under reduced pressure. The
distillate was then partially frozen and the mother
liquor decanted from the crystals. The crystals were
allowed to melt and this liquid was again partially
frozen and the mother liquor decanted from the crys-
tals. After several such treatments, in which the im-
purities in the original material are removed in the
liquid, a constant freezing point of — 10.5° was reached.
Frequently a supercooling of ONT to about — 16° was
observed before crystals appeared, after which the
thermometer rose to — 10.5°. On several occasions
another rise in temperature to — 4.45° was noticed,
accompanied by a crackling sound. The existence of
1 This paper is the fifth of a series dealing with the freezing points and
thermal properties of the nitrotoluenes, the investigation having been
undertaken at the request of the Division of Chemistry and Chemical
Technology of the National Research Council.
2 This Journal, 11 (1919), 1 124.
3 In the paper by Bell and Herty (page 1125) there is a discussion of
the various values for the melting point of MNT, many citations giving
54° while others are around 51.5°. We have recently found an explanation
of the discrepancy in an article by Holleman (Rec. trav. chim., 33 (1914),
5), who found a sample of the material originally used by van der Arend.
The melting point given by the latter, 54°, was the original of all the cita-
tions giving the higher value. From a redetermination of the melting point
with the same material as originally used, Holleman concludes that the pub-
lished value 54.4° is a misprint for 51.4°. This brings all the determinations
within a few tenths of a degree of agreement.
two freezing points indicates the existence of two dif-
ferent crystalline forms of ONT, an observation which
we found had already been made by several investiga-
tors.1
MELTING POINTS OF THE TWO FORMS OF ONT
The metastable form of ONT (a-ONT) always ap-
pears first, and frequently remains unchanged for sev-
eral hours even when the freezing liquid is stirred
vigorously. Where the stable form of ONT (/3-ONT)
was desired, von Ostromisslensky cooled the liquid to
— 50° or — 60 ° in solid carbon dioxide. At first the
metastable form appeared, but after a very short time
transition to the stable form took place with a crack-
ling sound. During our work a much simpler method
was found, based on an observation made in an at-
tempt to obtain the eutectic temperature for MNT
and a-ONT. All attempts to find this temperature
failed because of the change of metastable ONT to
the stable form. To get the stable form we seeded
liquid ONT at about — 10° with a few crystals from
the eutectic mixture above described. The tempera-
ture immediately rose to — 4-45° (corr.) and remained
constant to complete solidification. This material was
kept in a low-temperature bath for "seed" purposes.
is
10
MNT
ONT
These temperatures are very close to those found
by von Ostromisslensky: — ■10.56° and — 4.14°. The
earlier results, however, are more at variance with
these. Thus, von Schneider2 gives — 14.8°, and Lep-
sius, in a private communication to Knoevenagel, gives
1 von Ostromisslensky, Z. physik. Chem., 67 (1906), 341; Knoevenagel,
Ber., 40 (1907), 508; both of whom cite D. R. P. Kl. 120, No. 158,219.
1 Z. fhysik. Chem., 19 (1896), 157.
6o
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
- — 9. 40 and — 3-6°, the former figure being later revised
to — 8.95°.
The du Pont Company has kindly furnished us with
results on the binary system MNT-ONT, in which
the freezing point for MNT accords well with our de-
termination, but the freezing point for ONT is given
as — 3-3°- We are now unable to explain the rather
large difference between these results ranging from
— 3. 3° to — 4-45°. In our work we purified several
different lots of ONT by the method described above,
which is also the patented method cited above, and
obtained a constant freezing point unaltered by fur-
ther crystallizations.
BINARY SYSTEM: MNT nM1
The freezing points and compositions of the mix-
tures for this system are given in Table I and Fig. 1.
Table I — Binary System (j-Nitrotoi.usne-o-Nitrotoi.ubnk
Table II — Binary System: Dinitrotoluene-o-Nitrotoll
Per cent by
Weight
Freezing
MNT
ONT
Point
Solid Phase
0
KID
—4.45°;
10
90
8 !
0-ONT
20
80
—12.8 :
30
70
—6.13 |
40
60
6 !2
50
50
16.84 1
60
70
40
30
25 65
32 58
MNT
80
_'0
39 :: \
90
10
45.68
100
0
51,3 1
In the figure the points are observed to fall on two
curves, one representing mixtures from which MNT
is separating and the, other representing mixtures from
which /S-ONT is separating. The eutectic temper-
ature and composition are ■ — 15.73° and 26 per cent
MNT. We were able also to obtain one point on the
curve where a-ONT is the solid phase. This curve
begins at the freezing point for the metastable ONT
and is roughly parallel to the curve for the stable ONT.
In the diagram the unbroken lines represent conditions
which it was possible to attain, the unstable conditions
appearing as dotted lines.
A study of this system has already been made by
Holleman and Vermeulen,2 although their paper was
not found until the present work was completed. It
is interesting that they were able to follow to the
eutectic point the curve for a-ONT, and give for the
eutectic temperature — 20.6°. The unpublished re-
sults of the du Pont Company and the results of Holle-
man and Vermeulen are in general in close accord with
the present results. Our curve for MNT lies slightly
higher than the du Pont curve, which in turn is slightly
above the curve of Holleman and Vermeulen. The
three sets of results for the ONT curve also show dif-
ferences, as the curves cross at a slight angle. The
eutectic temperature is given as — 14. 6°, as — 15.73°,
and as — 16. 40, the first by Holleman and Vermeulen
and the last by the du Pont chart.
BINARY SYSTEM: DNT-ONT3
The data for this system are represented in Table
II and in Fig. 2. In this case, like the preceding
system, there are two curves crossing in a eutectic
point. The temperature and composition for the
! Experimental work by F. H. Spry.
' Rtc. Iras, chim., 33 (1914), 1.
1 Experimental work by E. B. Cordon.
Per cent by Weight
Freezing
DNT
ONT
Point
Solid Phase
0
100
— t.45°l
5.6
9.9
94.4
90.1
-6.2
—7.7
— 10.5 J
0-ONT
18.2
81.8
30
70
5.30 }
40
60
19.50
50
50
29.19
60
70
40
30
39.39
48.36
DNT
80
20
55.46
90
in
62.55
100
0
69.55
eutectic are — 11.45° and 21 per cent DNT. We were
able to follow the curve for the metastable ONT for
a short distance and have represented it by an un-
broken line in the figure, the continuation as a dotted
portion representing unstable conditions. The un-
broken portion of this line is plotted from two deter-
minations in which the metastable ONT was used as
seed and did not change over to the stable form before
the determination was complete.
DNT
ONT
BINARY SYSTEM: TNT-ONT1
The data for this system are given in Table III and
in Fig. 3. It was possible in this case to follow out
curves both for a-ONT and for 0-ONT to their respec-
tive eutectic points with TNT, the eutectic for TNT
1 Experimental work by W. White.
Jan.,
1921 THE JOURNAL OF
INDU
Ta
31.E III — Binary System:
Trinitrotoluene
-0-NlTROT<
Per cent by Weight
Freezing
TNT ONT
Point Solid Phase
0 100
— 4.45<\
4.77 95.23
-5.7
0-ONT
9.17 90.83
—6.85
15.28 84.72
—8.7
0 100
— 10.35
4.77 95.23
—12.00
■
a-ONT
9.17 90.83
—13.3
25 75
—0.2
30 70
10.2
40 60
25.7
50 50
37.1
60 40
47.4
TNT
70 30
56.5
80 20
65. 1
90 10
73.0
100 0
80.35 '
and 0-ONT falling at —9. 7° and 19.5 per cent TNT,
and the eutectic for TNT and a-ONT falling at— 15.6°
and 16 per cent TNT. In obtaining these freezing
points we used the seed of the stable ONT in every
mixture.
TNT
ONT
In this paper we have given the data for three binary
systems of the nitrotoluenes. one of these nitrotoluenes
having two crystal forms. In one case it was possible
to follow the freezing-point curve for the metastable
form right to the eutectic point.
61
THE PREPARATION AND ANALYSIS OF A CATTLE FOOD
CONSISTING OF HYDROLYZED SAWDUST1
By E. C. Sherrard and G. W. Blanco
Forest Products Laboratory, U. S. Department op Agriculture*
Madison, Wisconsin
Although the Forest Products Laboratory has con-
sidered for some time the advisability of invescigating
the nutritive value of hydrolyzed sawdust, it was not
until the severe drouth, which occurred last year in
the Northwest, called our attention to the pressing
need of such a material that the investigation was
undertaken. The product described in this paper
was prepared by this laboratory, and fed to three
dairy cows by the Wisconsin College of Agriculture
with highly gratifying results. While the experiment
is yet in the preliminary stages, it is deemed advisable
to describe the process of manufacture and present
the analysis of the original and digested sawdust.
PREPARATION OF MATERIAL
The sawdust was eastern white pine obtained from
a mill in Minnesota, and was representative of the
waste obtained from mills cutting this species. No ef-
fort was made to remove bark or other foreign sub-
stances that ordinarily are present in this material.
The sawdust was treated in the same way as for
the production of ethyl alcohol from wood; that is,
it was digested with 1.8 per cent sulfuric acid for 15
or 20 min. under a steam pressure of about 120 lbs.
per sq. in. Sufficient water was added along
with the sawdust to raise the ratio of water to dry
wood to about 1.251. After the steam pressure had
been blown off to atmospheric pressure, the treated
sawdust was removed from the digester, and a large
portion of the acid liquor removed by means of the
centrifuge. The centrifuged material was then placed
in towers, and the remainder of the sugar and sulfuric
acid extracted with hot water. The leach water was
mixed with the centrifuged liquor, and the whole
almost neutralized with calcium carbonate. After
the sludge had settled, the liquor was decanted or,
if necessary, filtered, and evaporated under reduced
pressure to the consistency of a thick sirup.
The leached material from the towers was screened
through a 6-mesh screen to remove the larger uncooked
pieces of wood, and the screenings dried by spreading
on the floor in a thin layer. The air-dried hydrolyzed
dust was then mixed with the sirup referred to above,
and the whole dried to about 1 2 per cent moisture.
Early in the experiment, when we were dependent
upon the air drying of the finished product, considerable
loss of sugar was experienced. For instance, in Cook
No. 139, 21.2 per cent of the dry weight of the original
wood was converted into sugar. The final wood meal,
however, contained only 16.39 per cent of sugar calcu-
lated upon the dry weight of the product. This
loss of almost 5 per cent sugar was partly due to the
mechanical treatment and partly to a slow fermentation
of the sugar in the moist product during the early
stages of drying. Table I shows the decrease of sugar
1 Presented before the Division of Industrial and Engineering Chem-
istry at the 60th Meeting of the American Chemical Society, Chicago,
111., September 6 to 10, 1920.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. i3, No. i
in samples containing over 12 per cent of moisture,
upon standing from 1 to 2 mo. at ordinary room tem-
perature.
Table I — Change in Sugar Content upon Drying
Date
7/3/19
7/3/19
7/3/19
7/21/19
7/31/19
8/15 '19
Moisture
Per cent
14.76
22.48
30.57
18.77
7.00
15.74
13.61
13.76
16.39
1 8 . 06
14.96
15.88
Date
9/22/19
, 1 1 lg
9/22/19
Moisture Sugar
Per cent Per cent
8.70 13.23
It will be noted that in the samples containing 15 per
cent or less of water, but little change in sugar con-
centration occurs upon standing. A gradual decrease
in sugar occurred in all samples that were air-dried.
0
''
'"
90
'
/
/
eo
/
/
' 1
1
1 J
1
1
j
1
P
1
/
1
0
)
fi
v
II
r
Ik
/
L
£
G
£
Nl
■>
1 J
5
\/f.
hi.
ri
■ >
1c
d
£
/
Tc
ra
/ A
ec
>«c
S
uq
Vr
/
/
//
/
/
'/
/
/
O
A/o of £xrracr/ons
Extraction ok Sugar and Sulfuric Acid
In order to overcome this difficulty, a drying oven
was installed and the moisture in both the leached
dust and final product reduced to less than 15 per
cent before storing. No loss in sugar has been noticed
in this material, even after storage of several months.
That the sugar content is but slightly lowered during
the drying is shown by Table III. The temperature
of the oven remained almost constant, but considerable
rise in temperature was noted in the dust. The
temperature of the latter was taken by a thermometer,
the bulb of which was covered with the drying ma-
terial.
During the course of the experiment it was found
desirable to determine the relative ease with which
the sugar and acid could be removed from the cen-
trifuged hydrolyzed wood. This was because of the
desirability of removing almost all of the sulfuric acid
and of leaching out as little sugar as possible. Under
ideal conditions a minimum quantity of water should
be used, thus lessening the volume to be evaporated
eventually.
Since it was also important that a complete analysis
be made of the original wood and the wood after
treatment, proportionate quantities of the centrifuged
dust and liquor were taken from Cook No. 164 and
the process completed on a laboratory scale, thus
avoiding some of the losses usually experienced in
working with large quantities.
LEACHING EXPERIMENT
In order to determine the quantity of water necessary
to remove the greater part of the sulfuric acid, 6.06
lbs. of the centrifuged digested sawdust, corresponding
to 2.81 lbs. of dry material, were placed in two per-
colators, and 2.81 lbs. of water added to the first.
The percolate was collected, weighed, and transferred
to the second percolator. The percolate from the
second percolator was again weighed and the acidity,
specific gravity, and sugar determined. It is re-
gretted that equal extraction periods were not used.
but because of the laboratory hours this was found to
be impracticable. The acidity is expressed in degrees,
and represents the number of cc. of 0.1 N sodium
hydroxide solution required to neutralize 10 cc. of
the extract.
The sugar was determined as dextrose by means
of the method recommended by the U. S. Bureau of
Chemistry1 with one or two minor modifications.
This method is briefly as follows:
The sugar solution is carefully neutralized with anhydrous
sodium carbonate and allowed to stand for about 3 hrs. The
precipitated material is filtered off, and the clear filtrate diluted
so that 25 cc. will contain not more than 0.250 g. of dextrose.
Thirty cc. of copper sulfate and 30 cc. of alkaline tartrate solution,
prepared according to AUihn's modification of Fehling's solu-
tion, are mixed in a 250 cc. beaker with 60 cc. of water, and heated
to boiling. Then 25 cc. (duplicate) of the solution to be examined
are added and the boiling is continued for 2 min., taking the
time when, one-half of the 25 cc. of solution has been added.
The precipitated cuprous oxide is readily filtered in a porcelain
Gooch crucible with asbestos pad, and washed thoroughly
with hot water without any effort to transfer the precipitate
to the filter. The cuprous oxide is dissolved in 1 to 1.5 cc. of
nitric acid (sp. gr. 1.42), the asbestos filtered off, and washed
thoroughly with hot water. The copper filtrate, which has been
diluted to approximately 225 cc, is warmed to 60 to 650 C,
and electrolyzed for 1.5 to 2 hrs., using a current density of
1.0 amp. per sq. dcm. of platinum gage cathode, and an e. m. f.
of 1 .6 volts. The cathode is removed while the generator is
still running, dipped into three changes of hot distilled water,
and finally washed with alcohol and ether. Afterward the
electrode is dried for 3 min. at 105° C, allowed to cool, and
weighed. From the amount of copper deposited the quantity
of reducing material can be calculated in terms of dextrose by
referring to Allihn's tables.
1 Bureau of Chemistry, Bulletin 107, 49.
J an . . 1021
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
T
4BI.E II— RE
UXTS
FROM
Leachino I
Experiment
Water Used ■ — Started .
Lbs. Pi P2
. -Fir
Pi
ished .
Pi
,-Ti
Pi
Hrs.
pT
Hrs.
Quantity-
Obtained
from Pi
Lbs.
Weight
Lbs.
-Obtained
f P
Per cent
Total
Sugar
Removed
Total
Sulfuric
Acid
Removed
Nc
Sp. Gr.
Reducing
Sii^ar
Per cent
I
2.81
3/26/20
11: 15 A.M.
3/26/20
3: 15 P.M.
3/26/20
3: 15 P.M
3/27/20
8: 15 A.M.
4
17
1.625
0.545
23.0°
1.032
5.37
9.40
10.34
II
2.81
3/26/20
3 : 20 P.M.
3/27/20
8: 25 A.M.
3/27/20
8: 15 AM
3/28/20
12 30 P.M
17
28
2.645
2 . 240
21 .2°
1 .029
5.01
36.02
37.64
III
2.81
3/27/20
8:40 A.M.
3/28/20
12:30 P.M.
3/28/20
[2:30 P.l
3/28/20
. 8: 30 A.M.
28
20
2.710
2.630
13.4°
1 018
3. 10
26. 15
27.86
IV
2.81
3/28/20
12:35 p.m.
3/29/20
8:30 a.m.
3/29/20
8:30 AM
3/30/20
8: 30 A.M.
20
24
2.755
2.715
6.9°
1 .010
1.69
15.02
15.23
V
Centrifuge
Liquor
2.81
3/29/20
9:00 A.M.
3/30/20
9:00 A.M.
3/30/20
8:30 A.M
3/31/20
8: 30 AM
24
24
2.760
2.725
2.850
3.1°
1 .004
0.71
0.33
6.25
i 0:
6.89
1 90
Original materia! 2.81 lbs. dry weight, contain
HiS04 : Vol. acid : : 4.2 : 3.
Pi — 1st Percolator : P2 — 2nd Percolator.
ng 1 1 .09 per
cent
total 1
educing sugar.
It will be noted from Table II and from the extraction
curves that all but 2.04 per cent of the total acid used is
removed by the fifth washing. Since only 1.8 per cent
sulfuric acid was used in the cook, there remains 0.026
per cent of acid in the finished stock food. The
liquor obtained by centrifuging the residue after the
final extraction contained 1.9 per cent of the total
sulfuric acid, so that in actual practice it would be
possible to remove practically all of the acid either
by centrifuging or by pressing. The sulfuric acid
concentrations used in the table and curves were
calculated from the total acidity using the ratio of
sulfuric acid to volatile acid as 4.2 : 3, as determined
by Kressman.1
The sugars were found to leach with a little more
difficulty, since 7.16 per cent of the total amount re-
mained in the residue after the fifth washing. This,
however, makes no difference in the final product,
since the sugar is not appreciably changed by
drying.
The liquor obtained from the extraction was com-
bined with the original digester or centrifuge liquor,
and the whole neutralized with dry calcium carbonate.
No change in the sugar concentration was noticeable
after neutralization. The mixed liquors were evap-
orated under reduced pressure to a thick sirup, and the
sirup mixed with the partially dried, extracted dust
which had previously been screened through a 6-mesh
sieve. The moist mixture was then placed in an oven
and dried. Although the per cent of total reducing
sugars decreased somewhat during the drying, the
fact that the total soluble solids remained almost
Table III — Analysis
Date
4/9
4/9
4/9
4/9
4/9
4/9
4/9
4/9
4/9
4/9
4/9
4/10
4/10
4/10
4/10
4/10
4/10
4/10
4/10
4/10
4/11
4/11
Hour
11:30 A.M
Noon
1 :30 p.m
2:00 P.M
2:30 P.M
3:00 P.M.
3:30
4:00 P.M
4:30 P.M
7:30 P.M
11:30 P.M
3:30 A.M
7:30 A.M
9:00 A.M
9:30 A.M
10:00 A M
1 1 :00 a m
Noon
6:00 P.M.
12:00 P.M
6:00 A.M
Noon
Temperature
Kiln Food Moisture
0 C. ° C. Per cent
75 Started 60.23
during Drying in Kiln
Total Ratio
Reducing Soluble Sugar
Sugars Solids Total to
Per cent Per cent Sol. Solids
18.63 26.15 71.3
50.38
45.59
40.93
35.50
17. 17
17.14
16.84
24 . 99
23 52
2,1 . 02
24.80
1 7 . 53
25.28
69.5
17.49
24.87
70.03
16.57
24.25
68.4
16.53
24.69
67.2
16.51
25.55
65.0
constant indicates that volatile reducing substances
were removed and that the sugar remained practically
unchanged. The progress of the drying experiment
may be observed from Table III.
After drying, the material contained considerable
finely powdered dust. The size of these particles was
roughly determined by screening.
Total weight of material
Material retained by 80-mesh
Material retained by 100-mesh
Material through a 100-mesh *
i = 499 g.
i = 13 g.
74.36 per cent
1 .93 per cent
22.50 per cent
Unpublished bulletin
Any loss of wood meal that occurs in handling con-
sists mostly of fine material, due to its sifting through
the bags or loosely made containers. It was therefore
analyzed separately, in order to determine its relative
value as compared with that of the coarser material.
The portion that passed through the ioo-mesh screen
was kept separate. The coarser material that was
retained by the ioo-mesh screen was ground to pass
through an 8o-mesh screen but to be retained by a
ioo-mesh. This was found to be impracticable,
owing to the fact that the coarse material ground itself
away on the screen and but little remained. Because
of this trouble, all of the coarse material was ground
to pass a ioo-mesh screen.
In this way two portions of the wood meal were
obtained: The portion that passed through the ioo-
mesh screen before grinding, labeled "unground food
through ioo mesh," and the ground portion labeled
"ground food through ioo mesh." For the purpose
of comparison a sample of the original white pine
sawdust was ground, and the portions passing through
8o-mesh and ioo-mesh screens were used for analysis.
It should be borne in mind that this analysis is not
comparable to the average wood analysis since no
effort was made to eliminate bark or other undesirable
portions of the wood. In fact the material used was
typical sawmill waste, and contained all the foreign
substances common to this product.
The two samples of stock food and the two samples
of unhydrolyzed sawdust were analyzed according to
A. W. Schorger's method.1
In both the untreated wood and the final product
the percentage of ash is higher in the fine material.
This is due possibly to the presence of sand and earth
that was contained in the original sawdust.
It will be noticed in examining the analytical data
in Table IV that the hot and cold water and alkali-
1 This Journal, 9 (1917), 556
64
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
Table IV — Analysis of Wood Meal
Cold
Moisture Water
80-100 6.00 8.81
-Solubility of Sample i
Untreated White Pine Sawdust
Pento
Sample
Unhydrolyzed dust
mesh
6.46
Average 6.23
Unhydrolyzed dust through 6.38
100-mesh
6.39
Average 6.39
Unground wood meal through 4.15
100-mesh
4.09
Average 4.12
Ground wood meal through 3.64
100-mesh
4. 1 1
4.13 23.16
8.75
9.21
10.08
10.15
10.78
22.86
23.01
25.39
30.17 30.64
30.59
30.38
31.11
30.69
30.66
28.23
4.84
4.86
3.94
■V. 77
3.85
3.80
25.76
25.57
43.24
42.59
42.92
40.23
Average
iround wood meal thr.
80-100 mesh
2.83 3.83
1 This value is undoubtedly somewhat low since the condenser was accidentally removed before the flask had sufficiently cooled.
- Pento- Cellu- Cellu- Cellu- Crude
sail lose lose lose Lignin Fiber
... 56.31-56.00 31.45
2.37 56.63-57.50 8.00 1.65 (30.65) 63.87
Re-
ducing
Sugars
56.61
53.76
54.11
36.01
35.97
35.99
37.77
Ash
0.82
6!80
0.81
1.52
4.92
4.93
3.35
37.90
37.46
37.23
soluble materials have been greatly increased by the
hydrolysis, while the ether-soluble remains about the
same.
In comparing the yield of pentosans from the original
wood and from the completed stock food, it will be
seen that considerable difference exists. Since the
yield of finished stock food is about 90 to 94 per cent
of the original wood, the pentosan yield from the
stock food amounts to about 4.05 and 4.43 per cent,
respectively, when calculated upon the dry weight of
the original wood. In other words, about 45.4 per
cent of the original pentosan remains in the finished
product. This difference is best accounted for by
assuming a partial conversion of the pentoses liberated
by hydrolysis into volatile acids and furfural.1 Such
an assumption is necessary to account for the volatile
acid formed in the condensed blow-off and centrifuged
liquor. Although but little difference is apparent
in the quantity of acetic acid obtained by the acid
hydrolysis of the original wood and treated wood,
too much confidence should not be placed in the quan-
tity of acetic acid obtained from the stock food, since
there is a possibility that a portion of this was liberated
from the calcium salts formed during neutralization.
The methyl pentosans in the wood are almost un-
affected by the digestion with sulfuric acid.
The average yield of cellulose in both samples of the
original wood is 55.79 per cent, while the average
yield from the stock food is 37.08 per cent. When the
cellulose from the latter is recalculated upon the original
dry weight of the wood the average yield is 34.11
per cent. This indicates a loss of 21.68 per cent of
cellulose from which 15.5 per cent of total reducing
Sugars were produced. The latter value, which is also
calculated from the original dry weight of the wood,
shows a yield of sugar corresponding to 71.5 per cent
of the theoretical, assuming that all of the cellulose
that is removed goes to form reducing sugar. The
calculation is at best an approximation, since the com-
plexity of the cellulose molecule, and hence the number
of molecules of water entering into the reaction, is not
known.
LIGNIN DETERMINATION
The method used for the lignin determination was
■ Kressman's unpublished bulletin.
that described by Mahood and Cable,1 except that a
16-hr. digestion with 72 per cent sulfuric acid was
used, since in a more recent study these authors found
the longer period more desirable.
The lignin determination is of interest, since it
shows that the total quantity of lignin contained in
the wood is not appreciably altered. The values
contained in parenthesis are the ash-free values cal-
culated from the dry weight of the original wood and
indicate that no change has occurred in what is
ordinarily considered as the lignin complex. This
is of great interest since heretofore the assumption
has always been made that a large portion of the lignin
was removed.
DETERMINATION OF a-, j3-, AND 7-CELLULOSE
The determination of a-, /?-, and 7-cellulose2 was
carried out as follows: About 2 g. of cellulose obtained
by the chlorination method were thoroughly mixed
with 20 cc. of 17.5 per cent sodium hydroxide and
allowed to stand for exactly 30 min. at room tem-
perature. The mercerized fiber was then treated
with 20 cc. of water, thoroughly stirred, and filtered
on an alundum crucible with the use of strong suction.
The a-cellulose which remained in the crucible was
washed with 10 cc. portions of cold water until the
filtrate showed no alkaline reaction. It was then
treated with hot 10 per cent acetic acid, washed six
or eight times with hot water, dried at 1050 C, and
weighed. The alkaline filtrate was made distinctly
acid with concentrated acetic acid, which caused
the /3-cellulose to separate in a finely divided condition,
and the brownish color of the liquor to become lighter.
To coagulate the suspended material, the solution
was heated in a water bath until the particles settled
and the solution became clear. The /3-cellulose was
then filtered on an alundum crucible, washed six or
eight times with hot water, dried at 105 °, and weighed.
The portion of the cellulose permanently dissolved was
7-cellulose.
No difficulty was experienced in the determination
of a-, /?-, and 7-cellulose in the cellulose obtained from
' Paper, 26, No. 24.
2 Cross and Bevan, "Researches on Cellulose," 1908-10, Vol. Ill, p. 23;
Cross and Bevan, "Paper Making," 1916, p. 97; Schwalbe, "Chemie der
Cellulose," 1911, p. 637.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
"
the unhydrolyzed sawdust, except in one or two cases
where the filtration was slow, owing to the porosity of
the crucible. The results in Table V were obtained
using the original untreated sawdust.
Table V — Per cent a-, 0-,
1 Original
Cellulose Sample Obtained from a-Cellulose tf-Cellulose 7-CeIlulose
Unhydrolyzed dust through 80-100
mesh 57.3o 19.61 23.03
Mixing cellulose obtained from un-
hydrolyzed sawdust. 80-100 mesh,
and unhydrolyzed sawdust through
100-mesh, respectively 55.85 29.42 14.75
In the case of cellulose obtained from the hydrolyzed
wood, considerable difficulty was encountered, owing
to its character after treatment with the alkali. In
all cases it was impossible to filter in the 30 min. pre-
scribed by the method, so that the action of the alkali
continued in some cases for 8 or 10 hrs. This difficulty
could not be overcome, and no definite analysis could
be made. The cellulose, upon treatment with alkali
(17.5 per cent), became semitransparent and had
the appearance of collodion.
That portion that could be drawn through the
crucible reprecipitated upon mild dilution with water.
This precipitate coagulated upon warming, and it
behaved and looked very much like the usual /?-
cellulose. The coagulated precipitate was filtered
on an alundum crucible with suction, and the filtrate
acidified with strong acetic acid, with the result that
no further precipitate was obtained. Because of the
difficulties outlined above, no analytical data on the
a-, f$-, and 7 -cellulose from the cellulose from hydrolyzed
wood are contained in this paper. It is hoped that
further investigations will clarify this point.
In one case the alkali-treated cellulose from hy-
drolyzed wood was strongly diluted with water.
The fine white precipitate was warmed, and the
coagulated material filtered, washed, and dried. It
had the semitransparent appearance of dried collodion
and amounted to 06 per cent of the original sample.
Because of its peculiar properties it is apparently a
product intermediate between a- and /3-cellulose.
Since it is partially soluble in alkali it may be con-
cluded that it is more easily digested in the alkaline
intestinal tract than the true a-cellulose, especially
in the presence of enzymes present in the intestines.
METHOD FOR CRUDE FIBER DETERMINATION
The crude fiber was determined ' by the method
outlined in Bureau of Chemistry Bulletin 107, page
56, with minor modifications. It is briefly as follows:
Two grams of the sample are extracted with ether for 4 or
5 hrs. in a Soxhlet extractor. The excess of ether is removed
by suction and the material dried to constant weight. It is
then treated with 200 cc. of boiling 1.25 per cent sulfuric acid,
and boiled under a reflux condenser for 30 min. After filtering
with suction on an alundum crucible it is washed with hot water
and treated with 200 cc. of boiling 1.25 per cent sodium hy-
droxide solution. After boiling for another 30 min. under a
reflux condenser it is rapidly filtered with suction through an
alundum crucible and washed with hot water until free from
alkali. After drying to constant weight it is incinerated in an
electric muffle at 7000 to 8oo° C. The loss on incineration is
considered to be crude fiber-
It is interesting to note that the crude fiber has been
reduced from 14 to 15 per cent. Another interesting
feature is the fact that the sum of the cellulose and
lignin is greater than the quantity of crude fiber.
This indicates that at least a portion of either the
cellulose or lignin, or perhaps some of each, is removed
by successive treatments with dilute acid and alkali.
SUMMARY
1 — A method for the preparation of a stock food
from white pine sawdust is described.
2 — Leaching experiments carried out on the digested
dust indicate that five complete washings with a
quantity of water equivalent to the weight of the wood
are necessary to remove the sulfuric acid. The
sugars were found to leach with somewhat more
difficulty than the acid.
3 — It is pointed out that the sugars contained in
the moist product are not appreciably affected by
drying at temperatures ranging from 75° to 85 ° C.
While some decrease is noted in total reducing sugars,
the loss is apparently due to the removal of volatile
reducing substances.
4 — A complete analysis is given for eastern white
pine sawdust,' and for the product obtained from the
same after digesting with dilute acid under pressure.
Attention is directed to the changes resulting from
this treatment.
5 — The cellulose obtained from the digested wood
differs from that from the original wood in its be-
havior toward alkali. In the former practically all
of the cellulose is converted into a viscous semi-
transparent mass by 17.5 per cent sodium hydroxide,
while in the latter over 50 per cent is unaffected.
THE EFFECT OF CONCENTRATION OF CHROME LIQUOR
UPON THE ADSORPTION OF ITS CONSTITUENTS
BY HIDE SUBSTANCE^
By Arthur W. Thomas and Margaret W. Kelly
Chemical Laboratories, Columbia University, New Yore, N. Y
The concentration factor in the combination of hide
substance with chromic oxide and sulfuric acid in
chrome liquor has previously been reported by Miss
M. E. Baldwin.2 She studied the adsorption from
various liquors containing 0.038 to 6.640 g. of chromic
oxide per 100 cc. of liquor, and found that the adsorp-
tion reached a maximum at concentrations of 1.5
to 2.0 g. of chromic oxide per 100 cc, beyond which
concentration the adsorption by the hide substance
decreased.
Results obtained by J. A. Wilson and E. A. Gallun3
in their investigation of the retardation of chrome
tanning by neutral salts, led them to believe that, had
Miss Baldwin's liquors been carried to higher concen-
trations (to about 12 g. of chromic oxide per 100 cc),
a minimum point might have been obtained beyond
which increasing concentration would have caused
1 Presented before the Leather Chemistry Division at the 60th Meet-
ing of the American Chemical Society, Chicago, 111., September 6 to 10,
1920.
' J. Am. Leather Chem. Assoc, 14 (1919), 433.
'-Ibid., 15 (1920), 273.
66
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13. No. 1
greater fixation of chrome. The experiments re-
ported in this paper were conducted to test this as-
sumption.
MATERIALS USED
The hide powder was American Standard (19 18)
of the same lot as used and analyzed by us.1
The chrome liquor contained 202 g. of chromic
oxide per liter. It was practically identical to that
used by Miss Baldwin. Eleven 200-cc. portions of
chrome liquor of various dilutions were made up from
this stock liquor.
METHOD
The various diluted liquors in 200-cc. portions were
poured into bottles containing 5.766 g. of hide powder,
equal to 5 g. of dry hide powder. Another portion of
each solution was set aside and at the expiration of 4S
hrs. the H+-ion concentration of the solutions was
determined. The bottles were shaken at intervals,
and at the end of 48 hrs. filtered off by suction. The
filtrates were set aside for analysis (the H+-ion con-
centrations determined immediately), and the chromed
hide powders, washed free of adhering liquor, were
air-dried. The methods of analysis were the same as
those reported by us in our earlier communications.
o
--rrr
=*-
^
^
1^,
t
"'--^-<
/
s'
^
■'-'
*
Original (a
fter 48 hours.)
The moisture was determined in each portion of the
chromed hide powders and all other figures calculated
to the water-free basis. The results are given in
Table I.
Table I — Composition of Chromed Hide Powder
G. CnOj per
100 Cc. of
Liquor before
Protein
Cr?Os
SO.
Ash
Adsorption
Per cent
Per cent
Per cent
Per cent
0.0363
98.19
1.30
1.09
1.59
0.2881
83.70
7.86
6.07
S.84
0.7738
76.63
10.58
8.18
11 .82
1.5526
75.90
10.85
8.67
12.12
3.0853
78.43
10.25
8.89
11.23
4.8073
80.17
9.36
8.25
10.09
7.3070
83.87
7.85
7.21
9.7267
84.83
5.92
6.12
6.50
12.175
89.77
3.86
5.19
4.89
14.754
90.67
2.35
4. 48
3.82
20.203
91.12
2.10
2.29
2.45
The analyses of the filtrates are given in Table II.
An aliquot part was taken in each case, the chromic
oxide in it determined and calculated to the basis of
100 cc. of liquor, assuming, erroneously, that no water
had been adsorbed by the hide — the common practice
in calculations of adsorption.
> J. Am. Leather Chem. Assoc., IS (1920), 487.
T\bi.r IT — Composition op Liquors apter Adsorption
N'timlier G. CnOi in 100 cc.
1 0.0096
2 0.0510
3 0.4464
4 1 . 2.SS6
5 2.8577
6 4.7587
7 7.4^0
8 10.0215
9 12.5820
10 15.4000
The H^-ion concentrations of the filtrates and of
the liquors (after 48 hrs.' standing) are to be found in
Table III and charted in Fig. 1. Those values which
are considered unreliable are in parentheses. In some of
the concentrated liquors we had difficulty in measuring
the H+-ion concentrations. The values obtained
show removal of hydrogen ion from the liquors up to
the solution of concentration of 7.4 g. chromic oxide
per 100 cc, beyond which the curves join and run
along together, indicating that if hydrogen ion was
removed the buffer action of the chromic sulfate
could take care of it. The solution which gave the
maximum adsorption of chrome in two days showed
a H+-ion concentration of 0.00056 mole per liter,
which checks Miss Baldwin's experience, where the
maximum adsorption of chrome in two days was
found to be from a solution of 0.0005 to 0.0006 mole
per liter concentration of hydrogen ion.
-Hydrogen-Ion Concentrations op Solutions
Filtrate from Liquor in
Contact with Hide Powder
for 48 Hrs.
Liquor after Standing
48 Hrs.
Mole per Liter of H +
0.00029
0.00039
(0.00060)
0.00056
0.00115
(0.00182)
0.00204
(0.00214)
0.00316
(0.00661)
Mole per Liter of H*
0.00004
0.00028
0.00042
0 . 00050
0.00083
(0.00110)
0.00186
0.00263
0.00316
(0.0045:
Table IV and Fig. 2 show the adsorption of chromic
oxide and sulfuric acid calculated to the basis of one
gram of dry hide substance.
Ms.
Cr-Oj ■
Ms SO,
From Analysis
From Analysis
From Analv
of Powder
of Liquor
of Powder
13.2
10.7
11.0
94.1
-
138.3
131.0
106.9
143.1
117.6
114 4
130.9
91.0
113.5
116.9
19.5
103. 1
93.7
—51.2
86.1
69.9
—118.0
72.3
43.1
—162.6
>, >)
26.0
— 25S.4
23.1
Solutions 3 and 4 showed the optimum concentra-
tion for a 2-day reaction with hide powder. The
chromed hide substance formed indicates a tetra-
chrome collagen, based on the equivalent weight of
collagen as 750, as suggested by Wilson.1 This again
checks Miss Baldwin's results quite closely.
The values based on analysis of the liquors, from
which the adsorption of water was ignored, show
lower values throughout, and from Solutions 7 to 1 1
J. Am. Leather Chem. As
12 (1917). 108.
Jan., 1921 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMIST R]
67
negative values are obtained, owing to the liquors
becoming more concentrated than they were originally,
on account of the collagen abstracting water from
them.
^ -210 —
-Crz03 from Analysis of Chromed Hide Ponder \
— ■— <T/> Oj from Analysis of Filtrate
Concentration of Liquor in Grams Cr20j per Liter
We would state our belief, based upon our experience
as presented in this and earlier papers, that the reaction
between chromic sulfate solutions and hide substance
is chemical and not physical, as contended by A. W.
Davison.1 If the adsorption were a simple physical
process, i. e., merely a partition of the chromic oxide
and sulfuric acid between the solid hide substance
phase and the solution phase, the curve should follow
Freundlich's adaptation of Henry's law: Ci = kC",
which is parabolic in shape; whereas Miss Baldwin's
and our experiments show that in a 2-day adsorption
the curve begins to slope steeply downward after the
concentration of the liquor exceeds approximately
16 g. of chromic oxide per liter in a solution of the com-
position of Cr(OH)S04, and reaches a minimum when
the concentration of chromic oxide is 147.5 S- Per liter,
this minimum being maintained at a concentration
of 202 g. per liter. This minimum confirms the pre-
diction of Wilson and Gallun in part. The most con-
centrated chrome liquor which we used was very
thick and about as concentrated as is possible to
handle; and therefore, we do not find it possible to test
further their prediction that increasing concentrations
beyond this minimum would cause greater fixation
of chrome.
ACKNOWLEDGMENTS
Acknowledgment is made of Mr. S. B. Foster's
assistance in the analytical work. We wish to express
our great appreciation of the generous support of
Messrs. A. F. Gallun and Sons Company in this
investigation.
J. Am. Leather Che
12 (1917). 258
THE ACTION OF CERTAIN ORGANIC ACCELERATORS IN
THE VULCANIZATION OF RUBBER— II'
By G. D. Kratz, A. H. Flower and B. J. Shapiro
The Falls Rubber Co., Cuyahoga Falls, Ohio
One of the early patents2 for the use of synthetic
nitrogenous organic substances in the vulcanization
of rubber refers to the dissociation constant of 1 X io~s
as the dividing line between accelerating and non-
accelerating bases. On the other hand, Peachey3
has pointed out that certain other substances which
are not basic, or but slightly so, are also exceedingly
active as accelerators. The number of examples in
this class, however, is relatively small.
In the course of the experimental work described in
this paper we have made a comparison of the sulfur
coefficients of a type mixture vulcanized with the as-
sistance of a number of accelerators closely related to
aniline and for which the dissociation constants are
known. We have also employed the hydrochlorides
of two of these substances, relatively weak and strong
bases, in order to observe the effect of the acid portion
during the vulcanization. The results obtained and
the conclusions drawn led us to employ the sulfides
of ammonia as accelerators and vulcanizing agents.
. Briefly summarizing these results, it was found
that with the substances tested there was apparently
no direct relationship between their dissociation con-
stants and their excess sulfur coefficients or physical
properties after vulcanization. In a closely related
series, such as aniline and its methyl derivatives, the
substance with the largest dissociation constant was
found to be the most active. However, the relative
activities of the members of this series were not pro-
portional to their dissociation constants. Generally
speaking, the activity of all of the substances could
be traced to the amino group, and depended to a large
extent upon whether or not substitution had taken
place in this group. In this respect, they should prob-
ably be regarded as substituted ammonias, rather
than as the more complex derivatives of other sub-
stances.
One effect of the basicity of two of the substances,
methylaniline and ^-toluidine, was determined with
the hydrochlorides of these two substances. Our
results showed that with substances of this type, the
first effect of the base is to neutralize the retarding
action of the acid formed in the decomposition of the
salt during vulcanization. We had previously sug-
gested this in a footnote in a former paper.4 We also
found that when the acid liberated in the decomposi-
tion of such a salt is neutralized by other substances
in the mixture, the activity of the hydrochloride is
very close to that of the free base. These results are
of particular interest, as Van Heurn5 has shown that,
whereas ammonium carbonate is moderately active
as an accelerator in a mixture of rubber and sulfur,
1 Presented before the Rubber Division at the 60th Meeting of the
American Chemical Society, Chicago, III., September 6 to 10, 1920.
2 D. R. P. 280,198 (1914).
3 J. Sac. Chem. Ind., 36 (1917), 950.
< Chem. & Met. Eng., 20 (1919), 420.
' Comm. of the Netherlands Government for Advising the Rubber
Trade and Industry, Part 6, 202.
68
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
ammonium chloride is inert. The former salt decom-
poses into ammonia and a weak acid, the latter into
ammonia and a strong acid, according to the following
reactions:
NHOsCOj — >
N'H.Cl
2NH3 + HjO + COa
— > NH3 + HC1
Our final experiments, wherein we found that in a
closed system rubber is vulcanized by heating with
ammonium polysulfide or ammonium hydrosulfide,
were carried on in order to obtain a reaction mix-
ture of undoubted basic character, which at the same
time would include H5S as one of the decomposition
products. The function of H2S in connection with
the vulcanization of rubber has long been made a sub-
ject of controversy. In the present instance it may
be regarded as a very weak acid.
Our results with ammonium polysulfide may be ex-
plained as due to the decomposition of this substance
into ammonia, hydrogen sulfide, and sulfur, the latter
substance being liberated in an active (nascent) form
which readily combines with the rubber. The analogy
between our results with ammonium polysulfide, and
those obtained years ago by Gerard1 with potassium
tri- and pentasulfides, is taken up in greater detail in
the experimental part of this paper. It is equally
evident, however, that if this explanation is advanced
in the case of ammonium polysulfide, vulcanization
with ammonium hydrosulfide requires that this sub-
stance decompose not into ammonia and hydrogen
sulfide only, but with the subsequent formation of a
polysulfide which liberates sulfur in the active form.-
It has been shown by Bedford and Scott3 that many
of the more complex substances which accelerate the
vulcanization of rubber react with sulfur, with the
liberation of H:S and the formation of thiourea deriva-
tives. In view of our results with the ammonium
sulfides, the action of such thiourea derivatives would
depend upon their ability to enter into a subsequent
reaction with the H;S formed, or the sulfur present in
the mixture, with the formation of a polysulfide.
Further, although the formation of a polysulfide in
this manner would, to a certain extent, be dependent
upon the basicity of the substance originally added
as the accelerator, it is obvious that the dissociation
constant of the reaction product would be a better in-
dication of its activity than the dissociation constant
of the original substance.
* R. Hoffer, "Treatise on Caoutchouc and Guttapercha" trans.
Brannt), H C. Baird & Co., London. 1883.
2 As an aqueous solution of XHtHS was employed, the action of this
substance may also be explained by its dissociation products. It would
dissociate with NHi* as the cation and HS~ the anion. As the HS~ion
itself is weakly acid, there would probably be many H+ and HS~ ions
and but few S ions in the aqueous solution. The H+ and S ions
in turn react to form H3S. On the other hand, (NHi)iS dissociates with
NH<+, the cation, and S , the anion. The latter, in the presence of
water, dissociates with the formation of OH" and HS" ions. Thus,
NH4HS dissociates with the formation of a greater number of H + ions than
in the case of (NHO:S, and consequently with a greater re-format:"-' of
H:S. This may account for the difference in the relative activities of the
two substances. The same may be true in the absence of water, as most
organic accelerators are apparently soluble in rubber, the high dielectric
constant of which indicates that this substance itself may be a good dis-
sociating medium.
' This Journal. 12 (1920). 31
In a previous paper1 we have suggested that the ac-
tivity of certain nitrogenous substances may be in-
terpreted on the basis of a change in valency of the
nitrogen, with the nitrogen functioning as a sulfur
carrier. This suggestion was made to assist in corre-
lating the nitrogen content with the activity of the
substances employed, although, as pointed out in the
above paper, results obtained by others already indi-
cated that the sulfur is not necessarily attached to
the nitrogen. While our present results show that
vulcanization may be effected by polysulfide forma-
tion, they do not exclude the possibility of the active
nitrogen group acting as a catalyst.
EXPERIMENTAL PART
The same general method of procedure was pursued
in the course of this work as was previously reported
in Part I.
The rubber was good quality, first latex, pale crepe,
and the same lot was employed for all mixtures. All
of the mixtures, the composition of which is shown at
the head of the various tables, were mixed and vul-
canized as before. Physical tests were made on a
Scott testing machine of the vertical type. Sulfur
estimations were made by our method, previously
described in detail.2
The accelerators were purified, and melted or boiled
at the temperature shown in the tables. All of the
accelerators were compared on a molecularly equiva-
lent basis, 0.01 g. molecule of the accelerator being
added for each ioo g. of rubber in the mixture.
expt. 1 — This experiment was carried on in
order to ascertain the relative accelerating effect of the
homologs of aniline and other closely related bases,
and also to compare the excess sulfur coefficients with
the dissociation constants of the substances originally
added as accelerators. The results obtained, together
with the physical constants of the substances em-
ployed as accelerators, are shown in Table I.
It is evident from this table that with aniline and
its methyl derivatives, or in the case of the two phenyl-
enediamines, the substance with the largest dissocia-
tion constant produces the greatest excess coefficient
of vulcanization. It is also apparent that this rela-
tionship is confined to more or less closely related
substances only, and that, as a general rule, the dis-
sociation constant is not a reliable guide to the ac-
tivity of a substance as an accelerator.3
Excess sulfur coefficients of equal magnitude (3.0)
were obtained from />-toluidine, ^-benzidine and m-
phenylenediamine. It is interesting to note that
the subtraction of the excess sulfur coefficient of any
one of these substances from that obtained for />-phenyl-
enediamine (5.2) leaves a figure very close to the
excess obtained for aniline (2.4"). Further, although
the mixtures vulcanized with the assistance of the three
substances in question were found to have the same
excess sulfur coefficient, all of them had widely differ-
1 This Jovrnal, 12 (1920). 317.
•■India Rubber World, 61 (1920), 356.
8 The dissociation constants given in Table I are taken from the Landolt-
Bornstein tables and are not strictly comparable, in that they were not all
determined by the same method.
Jan.. iQ2i THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
60
Substance Formula
Control
Aniline CtH.NH?
Methylaniline CH1NH.CH1
Dimethylaniline CsH.N(CHi).
/.-Toluidine CHj.C.H..NHi
ra-Phenylcnediaminc NHj.CsIft.NHj (1: 3
p-Phenylenediamine NHj.CeH«.NH2 (1 : '
^-Benzidine NH2.C«H1.CtH<.NH:
Phenylhydrazine C1H1NH.NH,
Hydrazobenzene* CtHsNH.NH.CsHs
All
■ belov
nd b. p. abo
Tabu! I
Sulfur
Vulcanized for 90 Min. at 148
0 C.
Determined
M. P. or B. P.
Dissociation
Excess
Strength
Constant K
Sulfur
Accelerator1
at 15° to 18° C
Coefficient
at Break
at Break
(2.581)
1229
1 83 . 1
3.50 X 10-"
2.400
2005
192. 0
2.55 X 10-'°
0.612
1665
192.5
2.42 X I0-i«
0.250
1938
1.60 X 10-»
2.987
2476
62.6
1.35 X 10-1=
2.986
1933
140.6
2.48 X 10-1= !
5.248
193
126.2
7.40 X 10-u >
3.056
1464
240.0
1.60 X 10-»
0.751
1052
126.0
0.777
2165
1140
ture of vulcaniz
ation. 3 Figure a
pplies to second "K "
3 Does not have basic properties.
ent physical properties. These substances may be
regarded as aniline in which hydrogen of the benzene
ring has been replaced by radicals.
On the other hand, (methylaniline), phenylhydra-
zine, and hydrazobenzene may be regarded as aniline
in which the hydrogen of the amino group has been re-
placed. The difference in the activity of these two
types of accelerators has already been mentioned in
Part I in connection with the phenylguanidines.
As in the previous instance, the same excess coeffi-
cient was obtained for (methylaniline), phenylhydra-
zine and hydrazobenzene, but the value was much
smaller (0.75) than before. The excess value found
for these three substances when subtracted from that ob-
tained for ^-phenylenediamine gives a figure equal to
about twice that obtained with aniline. Here, also,
the physical properties of the three mixtures were
greatly different.
I 1
*
1-
1 !-•
.-"Usis:
-.«
„.
'^M
■~t3?S*
irs^
**j"*
,•■■
t-i
r.
1
.
y
*
>--
^SiS-
13^-P1
r
/
3^S3
JgWafi^4-
/
7
A
/
k-
'
tensile strength in lbs per sq iv.
Fig 1
The discrepancy in the physical properties of mix-
tures vulcanized to the same sulfur coefficient by
means of different accelerators is of especial interest
and has been made the subject of a subsequent paper.
As our present results are based on one cure only, we
are not warranted in drawing many conclusions from
those recorded here. A comparison of the results
given in Table I, with the stress-strain curves shown
in Fig. 1, however, shows that these differences are
most evident at, or near, the point of break.1
The above results indicate that, irrespective of
whether or not an interaction between the accelera-
tor and other substances in the mixture takes place
during vulcanization, the activity of substances of
the type described is directly traceable to the amino
group, and particularly to the first amino group in
the benzene nucleus.
expt. 11 — In view of the results of Expt. I, they
should be analogous to those of ammonia or ammonium
salts. From a consideration of the work of Van
Heurn1 it seemed possible that certain other substances,
or their reaction products, active as accelerators, might
decompose with the formation of a (relatively) strong
base and a weak acid in an analogous manner; or that
some substances, which are not ordinarily classed as
accelerators, owing to their decomposition into a weak
base and strong acid, might be active if the acid so
formed was neutralized by another constituent of the
mixture. Aniline sulfate and />-toluidine hydro-
chloride, when employed in the presence of zinc oxide
are examples of the latter type.
Ta
BLE II
100
8. 1
Sulfur
8.1
r = 0.01 g. Mo
. of Substance
Vulcan
zed for 90 Min. at
148° C.
Physical
> — Properties—.
Sulfur
Sulfur
Tensile
Final
Coeffi-
Coeffi-
Strength Length
cient
cient
Lbs. per
Per
Sulfur
Over
Under
Sq. in.
Coeffi-
Control
Control
(At
(At
Mixture
cient
( + )
(— )
Break)
Break)
Rubber-Sulfur Control.
2.789
1265
1140
Zinc Oxide Control . .
~>.538
0.251
R-S Control + HCI
0.652
2.137
564
1250
ZnO Control + HCI
2.491
0.047
1783
820
5 . 568
2.779
2476
920
p-Toluidine + ZnO
5.371
2.833
1824
640
fi-Toluidine Hydrochloride
2.308
0.481
1070
1210
/>-Toluidine Hydrochloride
+ ZnO
3.990
1.460
2485
757
Methylaniline
3.193
0.404
1665
1050
Methylaniline + Zull
2.750
0.217
2237
800
Methylaniline Hydrochlo-
ride
1.012
1.777
530
1150
Methylaniline Hydrochlo
ride -f- ZnO
2.012
0.526
1731
840
i />-Phenylenedian:
suits could not be obta
so greatly
that concordant
for this substance.
In Table II are given the results obtained with
molecularly equivalent quantities of ^-toluidine and
/>-toluidine hydrochloride, in the presence and
absence of zinc oxide. Two control mixtures
were employed, one with and the other without zinc
oxide, no accelerator being added in either case. The
excess sulfur coefficients (+ or — ) shown in the third
and fourth columns of this table were obtained by the
subtraction of the coefficients of their respective con-
trols, depending upon whether or not they contained
zinc oxide.
From this tabic it is evident that zinc oxide itself ex-
erts a slight retarding action and that hydrochloric acid
THE JOURNAL 01 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
is a most effectual retardant, when employed in the ab-
sence of zinc oxide. When these two substances were
both present in the mixture, however, the sulfur
coefficient obtained was practically that of its control.
With />-toluidine, the same excess coefficient was
obtained in the presence and absence of zinc oxide,
a characteristic similar to that noted in the case of
aniline, and to be discussed in a subsequent paper.
In fact, />-toluidine hydrochloride did not greatly
retard the vulcanization even when employed in a
rubber-sulfur mixture, a fact which we attribute to
the strong basic nature of the />-toluidine. When
used in the presence of zinc oxide, ^-toluidine hydro-
chloride markedly accelerated the vulcanization. The
physical test results confirmed the sulfur coefficients.
Entirely different results were obtained, however,
with methylaniline hydrochloride.1 This substance,
although almost inactive in a mixture which con-
tained zinc oxide, acted as a retardant in a mixture
of rubber and sulfur only. In this instance, owing
to the weakly basic nature of the methylaniline. the
effect of the hydrochloric acid predominated. Here,
again, the physical properties of the mixtures were
roughly in accord with their sulfur coefficients.
The results show that the tendency of certain sub-
stances to decompose or dissociate into other sub-
stances with acid properties, or with acid properties
predominating, may cause the substance originally
added to be classed as inactive or as a retardant. In
such cases, the primary function of zinc oxide is to
neutralize the acidic constituents and permit the pre-
dominance of the accelerator, which is very probably
basic.
expt. in — Many years ago, Gerard2 noted that
vulcanization could be effected by boiling rubber in a
concentrated aqueous solution of "liver of sulfur," a
reaction which may possibly be represented in the
following manner:
4K2CO, + S10 • — > K:SO. + 3K2S-, + 4CO,
K0S3 + H,0 — *- 2KOH + H,S + S
The second reaction, which represents that found
by Gerard capable of effecting vulcanization, is analo-
gous to the decomposition of ammonium polysulfide:
(NH,):SX — > 2NH3 + H2S + Sx-i
In neither of the above instances is the possibility
of the formation of the hydrosulfide (KSH or NH«SH)
excluded, but it is regarded as an intermediate reac-
tion.
In the present case, where the ammonium sulfides3
were used, the resultant system can hardly be acid, no
matter how the decomposition or dissociation of the
sulfide is effected.
1 When heated to 350° C. methylaniline hydrochloride dissociates
into aniline and methyl halide, with the formation of the isomeric p-toluidine.
Methylaniline hydrochloride was chosen for comparison with p-toluidine
hydrochloride, in order to observe if such a rearrangement took place during
the vulcanization reaction. From the sulfur coefficients obtained, it is
obvious that this transformation did not occur.
! Loc. tit. The first use of alkaline sulfides, and particularly potassium
pentasulfide, for vulcanization, is often attributed to Gerard (compare
Charles Hancock, Brit. Patent 11,874 (1847), and Moulton, Brit. Patent
13,721 (1851))
s Compare the process of Moureley of Manchester, England, 1884.
A small sample of the rubber was sheeted thin on
the mill and cut into two 5-g. portions. Each portion
was placed in a glass bomb tube, and a concentrated
aqueous solution of ammonium polysulfide was added
to one, and ammonium hydrosulfide to the other.
Each solution contained approximately o. 5 g. of sulfide
sulfur. The tubes were sealed and heated for 6 hrs.
in an oil bath of 147 ° C.
Both samples appeared to be vulcanized to a slight
extent. The sample heated with ammonium poly-
sulfide was dark in color and quite sticky. The other
was lighter in color and not so sticky. Both samples
were extracted with acetone for 24 hrs., dried, and the
combined sulfur estimated.
The samples heated with ammonium polysulfide
and ammonium hydrosulfide were found to have sulfur
coefficients of 1.033 and 4-366, respectively.
CONCLUSIONS
1 — The activity of synthetic nitrogenous organic-
substances as accelerators is not proportional to the
dissociation constants of the original substances and.
with the exception of members of a closely related
series, no definite relationship exists between the activi-
ties and the dissociation constants of the original
substances.
2 — Substances which decompose or dissociate into
other substances of acid character, or react with other
components of the mixture to form substances of acid
character, do not accelerate unless a neutralizing
base or salt is present.
3 — Vulcanization is effected by heating rubber in
a closed system with concentrated aqueous solution of
ammonium sulfides.
ELECTRIC OVEN FOR RAPID MOISTURE TESTS
By Guilford L. Spencer
The Cuban-American* Sugar Co., New York and Cuba
The appreciation of the role of the moisture of raw
sugars in determining their storage qualities, and the
need of very prompt results of moisture tests in sugar-
cane bagasse, in controlling the mill work, led the
author to devise an oven for rapid tests. The ordinary
types of ovens are of great value in these tests, but
unfortunately the results in their use cannot be re-
ported with sufficient promptness to meet the needs
of thorough factory control. If raw sugar contains
more moisture than a certain safety factor indicates
is desirable, it may break down before it reaches the
market or refiner and serious loss of sucrose result. If
the residue of cane milling, the bagasse, contains ex-
cessive moisture, this necessitates a waste of fuel and
a loss of sugar.
The oven here described is the result of several
years' experimenting and the construction of several
models. As indicative of the rapidity that has been
achieved in the present model, raw sugar may be dried
in it in 10 min., and cane bagasse in about 30 min.
It was hoped to present comparative tests of several
materials and more systematic experiments with sugars.
' U. S. Patent 1.348,757.
1 Presented before the Section of Sugar Chemistry at the 60th Meeting,
of the American Chemical Society. Chicago, 111., September 6 to 10, 1920
Jan., 1 02 1
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
but the pressure of manufacturing duties prevented
the laboratories from doing more extended experiment-
ing.
DESCRIPTION OF OVEN
Briefly, the oven is a convenient device for convey-
ing a large volume of heated air through a capsule
containing the material to be dried. The current of
air is induced by a steam ejector or an air pump.
Connection is made with the vacuum system in sugar
factories. The heated air is carried against the cover
of the oven to promote mixing. The drying capsule
may be of metal or other construction, but, for sugar
work and a large proportion of general tests, metal
is most suitable. The bottom of the capsule is closed
with monel metal filter cloth, which freely passes air
but retains very fine powders. The capsule makes a
joint with its seat in the oven, over an annular channel
which connects several capsule openings, and leads to
the vacuum pump or ejector connection. The air
inlet to the oven may be regulated, if desired, for
operating under a partial vacuum. The air is drawn
over a heating element consisting of spiraled resis-
tance wire wound over a core. The travel of the air
is directed through a very narrow annular space, occu-
pied by the resistance wire, which forces it into inti-
mate contact with the resistor. The element is housed
inside the oven's drying chamber, thus reducing radia-
tion loss. The air pressure on the material in the
capsule forces the latter to a good seat and prevents
air leakage. The oven is made in two sizes, small
for general use and large for bulky materials.
The service wires are connected in series with a
sliding contact rheostat for temperature control, an
electric time switch or interval timer, and the heating
element. The time switch opens the circuit and rings
a bell at the termination of the drying period. The
heating element is housed conveniently for renewal.
OPERATION OF OVEN
The time switch is adjusted for the desired drying
period; the capsule, with the sample, is placed on its
seat in the oven and the unused openings are closed;
the vacuum or pump connection is opened; the time
switch is closed and the clock is started; the resistance
is rapidly cut out with the rheostat slide, and the tem-
perature is regulated. The drying now proceeds until
the time switch opens the circuit and rings a bell, sig-
naling the termination of the operation.
Any material that will freely pass a current of air
may be dried in this oven. Refinery press-cake, con-
sisting almost entirely of kieselguhr ("filter-eel") is
successfully dried. Liquids must be absorbed by a
suitable carrier and this be placed in a capsule. The
thermometer bulb must be located immediately over
the capsule. Owing to the short drying period, it has
not been found necessary to use a thermostat, though
provision is made for one. About one minute is re-
quired to heat the oven to the drying temperature.
The following experiment with absorbent cotton
indicates the rate of drying that may be attained: a
sample of cotton was dried to constant weight, then
saturated with distilled water, and in this condition
placed in a capsule in the cold oven and heated at
105 ° C.
Dry weight of cotton 0.8888
Weight after 5 min. drying 0.9858
Weight after an additional 5 min. drying 0.8889
COMPARATIVE TESTS WITH OLD TYPE OVEN
At intervals, this company distributes control sam-
ples among its laboratories, through its central control
laboratory. These samples are tested independently
by the chemists conducting the routine factory control,
and the results are reported to the author's office for
tabulation and comparison. The figures quoted below
are from such tests. A number of individual tests are
given to call attention to variations. In the tests of
raw sugar (Series I), the drying period was 20 min.
at 1050 C. with the new type of oven, starting with
the oven cold. In the usual types of electric oven, the
drying period was the customary 3.5 hrs. at 1050 C.
Series I
New Oven
Chemist A B C D' E1 F
Per cent moisture . .... . 0.72 0.73 0.72 0.78 0.78 0.70
Average per cent moisture = 0.74
Usual Type Electric Oven
Chemist G H I J K Control Laboratory
Per cent moisture . 0.68 0.74 0.69 0.72 0.75 0.76 0.79 0.77
Average of factory laboratories = 0.72
Average of control laboratory =0.77
Average of all tests =0.74
1 Kflluent air temperature, 95° C.
A second sample was sent to the various factory
laboratories, in which every precaution was observed
to assure thorough mixing of the sugar, and complete
filling and proper sealing of the bottles. A sugar of
very high moisture test was purposely selected. Four
heating periods were specified for the new oven, a
capsule of sugar for each, and the customary period
of 3.5 hrs. for the ordinary oven. The temperature
in each test was 1050 C. The results are tabulated
in Series II:
Chemist
Drying period, min. 3
Per cent moisture . . 1.36
Chemist .
Drying period, min. 3
Per cent moisture. . 1 . 28
Average = 1.45 (20 min )
H
SERIES II
New Oven
5 15 20
1.40 1.45 1.47
. A (retests)—
3 5 15
1.35 1.40 1.45
1.33 1.39 1.42
:ual Type Electric Over
1.38 1.42 1.44 1.44
Chemist
Per cent moisture. . 1.52 1.43 1.52
Average of 17 factory tests = 1 .48
' Effluent air temperature, 95" C.
1 . 43 1 . 50
ntrol Chemists
Lv., 1.50
The tests by Chemists D and E were made in an
early model of the oven in which the heating element
is immediately over the capsule. For this reason the
temperature of the air after passing through the cap-
sule is given. There is always danger of overheating
with this arrangement and it has been abandoned.
Most of these tests, except in the central control
laboratory, were made by young men with very little
laboratory experience. This applies to both ovens, so
these conditions were alike. Apparently the condi-
tions that lead to irregularities are no more in evidence
in the new than in the usual ovens. There is prob-
ably less danger of decomposition of the material dur-
ing desiccation in the new than in other ovens.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, Xo. 1
by reason of the very short heating period and the
prompt removal of the vapors.
Cane bagasse apparently withstands high tempera-
tures and is usually dried in the ordinary ovens at
no" to 115° C. It may be dried in the new oven at
130° or even 1400 C. without decomposition that in-
troduces an appreciable error. A sample weighing 100
g. and containing 50 per cent moisture may be dried
in the large oven at 130° C. in 30 min., the drying
period depending somewhat upon the mechanical state
of the material. Samples have been dried at the high
temperature, during various periods ranging from 3c
min to 00 min., without increase in the indicated
moisture.
ADDRL55E5 AND CONTRIBUTED ARTICLES
THE CHEMISTRY OF VITAMINES1
By Atherton Seidell
Hygienic Laboratory, U S. Public Health Service, Washington, D. C
The first indication of the existence of the substances now
designated by the term vitamine was obtained some twelve
years ago during the investigation of the cause of beri beri,
a disease prevalent among people who consume rice as their
chief article of diet. This disease originated after the intro-
duction of modern milling methods in which the surface layers
of the rice are removed by a polishing process. It was found
that the disease could be prevented by adding to the diet rice
polishings or extracts of these.
In 191 1 Casimir Funk, who was engaged in attempts to isolate,
by chemical means, the constituent of rice polishings responsible
for the remarkable curative effects, proposed that this hitherto
unrecognized substance be called vitamine. He also developed
the conception of deficiency diseases and collected much evidence
to prove that the absence of these previously unrecognized sub-
stances from an otherwise adequate diet is the cause of serious
nutritional disturbances, resulting in characteristic abnormal
conditions. Among such diseases he included beri beri, poly-
neuritis in pigeons, scurvy, and pellagra. The term vitamine,
therefore, refers to one or more substances of unknown composi-
tion, extremely small amounts of which are necessary for normal
nutrition.
Although many attempts have been made to isolate vitamine,
none have so far been successful, and our knowledge of this
class of substances is, therefore, still limited almost entirely to
the physiological effects they produce.
Since it has not been possible to determine the vitamine con-
tent of foods by chemical methods, feeding experiments for
this purpose have been developed and extensively applied.
The principle on which these are based is the feeding of diets
which contain adequate amounts of the hitherto recognized
essential dietary constituents, namely, carbohydrates, protein,
fats, and inorganic salts, highly purified to insure that they con-
tain no vitamine, and simultaneously giving measured amounts
of the sample being tested for its vitamine content. On the
basis of such experiments tables have been constructed which
show the comparative amount of vitamine in a large number of
foodstuffs. Furthermore, this work has led to the differentia-
tion of at least three well-characterized vitamines. These are
the water-soluble antineuritic vitamine, the fat-soluble, growth-
promoting vitamine, and the antiscorbutic vitamine. Of these,
the first appears to be the most stable towards the chemical
manipulations required for its separation from the substances
with which it occurs naturally. It is this one, therefore, which
has received most attention at the hands of chemists. Although
the results which have been obtained so far have not greatly
clarified the problem as to the chemical nature of this unknown
essential dietary constituent, it is believed that a brief review
of the experiments along this line may prove of general
interest.
1 Address of the retiring president of the Chemical Society of Wash-
ington, November 11, 1920.
EXPERIMENTAL PROCEDURES
At the cime Funk began work on the problem the following
facts had been qualitatively established in regard to the anti-
neuritic vitamine. It is neither a salt nor a protein. It i>-
soluble in water and in alcohol. It is dialyzable. and is destroyed
by heating to 1300 C.
Funk and others have since shown that it is not destroyed by
hydrolysis for 24 hrs. with 20 per cent sulfuric acid. It has
also been found that phosphotungstic acid precipitates this
vitamine completely from aqueous solution. Funk's method
for its isolation is, accordingly, based upon the use of this re-
agent. In general, the procedure consists in extracting the raw
material with acidified alcohol, evaporating the extract to a small
volume, acidifying the aqueous solution with about 10 per cent
of sulfuric acid, and precipitating with phosphotungstic acid.
This precipitate is decomposed with excess of barium hydroxide,
and after removal of the excess of the latter, the solution is
acidified with hydrochloric acid and evaporated. The residue
is extracted with alcohol and the alcoholic solution further puri-
fied by precipitating with various reagents, such as lead acetate,
mercuric chloride, silver nitrate alone and followed by barium
hydroxide, phosphotungstic acid, silicotungstic acid, etc.
Funk at first reported that the crystalline material he suc-
ceeded in isolating from rice polishings, yeast, milk, bran, and
other materials, by means of phosphotungstic acid precipitation
and subsequent decomposition of this precipitate, was the anti-
neuritic vitamine. Later, in collaboration with Drummond he
was forced to abandon this position since the compound he
originally thought was vitamine proved to be nearly pure nicotinic
acid. Retraction was therefore made of the claim that isola-
tion of the curative substance had been effected.
A number of other investigators have followed this general pro-
cedure and have reported the isolation of crystalline compounds
with antineuritic properties. Thus, Suzuki, Shimamora, and
Odake have given the name oryzanin to an active product they
obtained from rice polishings by alcoholic extraction followed
by phosphotungstic acid precipitation. Their experiments
were repeated by Drummond and Funk but their results were
not confirmed. Edie and his co-workers isolated a crystalline!
product from yeast by methyl alcohol extraction and silverll
nitrate baryta precipitation to which they gave the name
loruliti. but for which further evidence is lacking that it is pure
vitamine
Numerous modifications of the general plan of extracting and
precipitating have been tried without success and many novel
procedures have been introduced. Thus Sugiura recently made
use of air dialysis to obtain crystalline vitamine from water
extracts of dried yeast. The yield was very minute and physio-
logical tests of the product did not indicate that it possessed an
exceptionally high degree of activity. McCollum reported that
although organic solvents, such as ether, benzene, and acetone,
do not extract the antineuritic vitamine directly, if the alcohol
extract of the vitamine-containing material is evaporated on dex-
trin, and this extracted with the organic solvent, benzene ap-
pears to dissolve the vitamine, but acetone does so to only a
very slight extent.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
73
Recently, Osborne and Wakeman have proposed a modifica-
tion by which it appears that the removal of a considerable
amount of nonvitamine material can be effected by a very simple
expedient. This consists in adding the fresh yeast to slightly
acidified boiling water and continuing to boil this mixture for
about 5 min. This coagulates the protein and permits its com-
plete removal by filtration. The protein-free filtrate appears
to contain all of the vitamine originally present in the yeast.
An attempt to precipitate the vitamine fractionally from the
evaporated filtrate by means of increasing concentrations of
added alcohol was, however, only partially successful.
A procedure which has been found to offer marked advantages
in separating vitamine from the major part of the substances
with which it occurs in natural products is based upon
the property of being selectively adsorbed by certain
varieties of fuller's earth The particular variety found to be
most useful in this respect is that obtained from Surry, England.
In the case of brewer's yeast, which is, perhaps, the raw ma-
terial so far used to greatest extent as a source of vitamine, the
fresh yeast is permitted to antolyze, and the thick liquid which
results is filtered. This clear red-brown filtrate contains about
23 per cent of solids and is very rich in vitamine. If fuller's
earth is added to it in the proportion of 50 g. per liter and kept
in intimate contact with the liquid for about one-half hour and
then removed by filtration, the yeast liquor is found to contain
practically the same 23 per cent of solids originally present, but
all of the vitamine is now firmly attached to the fuller's earth,
and repeated washing does not remove an appreciable amount of
vitamine from it.
This fuller's earth-vitamine combination has, for convenience,
been designated as "vitamine activated fuller's earth." Physio-
logical experiments have shown that no noticeable deterioration
occurred in samples of the "activated solid" kept for over 2
yrs. Large amounts of it can be readily accumulated and,
after being uniformly mixed, it can be standardized by physio-
logical tests for its vitamine content. Such material forms a
particularly satisfactory starting point for the comparative study
of various methods for the isolation of vitamine.
In order to remove the vitamine from its combination with
fuller's earth, the only plan so far devised is based upon the use
of dilute alkali. This is a serious disadvantage since vitamine is
particularly unstable in an alkaline medium. It is, therefore,
necessary to operate rapidly and return to neutral or acid con-
dition promptly. The aqueous solution thus obtained from
"activated fuller's earth" has been found by physiological tests
to contain only about one-half of the total vitamine originally
present in the solid. There is every reason to believe, however,
that aqueous solutions so obtained are as free from extraneous
material as it has been possible to obtain in any other way.
Tests of the stability of the vitamine contained in them, made
by passing in air or oxygen, showed that comparatively little
destruction resulted. It is, however, not known how long the
vitamine activity is retained by such solutions.
Using the aqueous vitamine solution prepared as just described,
various attempts have been made to recover from it the active
material in the pure solid state. These attempts have so far
been unsuccessful. By careful evaporation of the solution, the
products successively obtained show more or less activity by
physiological tests, but in no case does the resulting material
possess the appearance or character which a pure product would
be expected to show. The action of solvents such as benzene,
acetone, ethyl acetate, and chloroform on these residues fails
to effect a separation of active from inactive material.
The numerous experiments which have been made with these
comparatively pure vitamine solutions have shown that the
vitamine tends to divide itself between the several fractions ob-
tained, rather than to become concentrated in one or the other.
The experiments are, however, always attended with con-
siderable uncertainty on account of the difficulty of keeping
track quantitatively of the vitamine. The only tests avail-
able for this purpose are feeding experiments, and even the sim-
plest of these require several weeks and give very uncertain results.
PHYSIOLOGICAL TESTS
The physiological test used by Funk and others of the earlier
workers was the cure of polyneuritic pigeons. By this test the
birds were fed exclusively on rice until they developed typical
paralysis, which ordinarily occurred within 2 to 3 wks. They
were then given measured doses of the sample in question. If
this contained vitamine, a remarkable improvement in the con-
dition of the pigeon occurred within a few hours. The diffi-
culty, however, is that a great variety of compounds may cause
an improvement:, and in some cases a temporary alleviation of
the condition may occur spontaneously. It is, therefore, very
difficult to interpret the indications of the test, and erroneous
conclusions may easily be drawn from it. The use of this test,
no doubt, accounts for many of the unconfirmed claims and con-
clusions which have been published in regard to the isolation
of vitamine. This curative test has now been abandoned by
almost everyone engaged in efforts to isolate vitamine.
The physiological method which appears to yield the most
trustworthy indications, as to the amount of vitamine present
in a given sample, may be referred to as the protective method.
A pigeon is fed exclusively on polished rice and simultaneously
given measured doses of the sample containing the unknown
amount of vitamine. It is weighed at frequent intervals and
if no loss in weight occurs within 2 or 3 wks., it is apparent that
the sample in question is furnishing an adequate supply of vita-
mine to meet the needs of the pigeon. If the amount of vita-
mine supplied is insufficient, a characteristic curve of loss in
weight will be obtained. It is apparent, however, that this
test will fail to show whether the sample contains more vitamine
than is just required to maintain constant weight. Hence,
quantitative results often require repetition of the test, and,
therefore, call for expenditure of much time and patience.
There is, consequently, a very great need for a more rapid,
accurate, and trustworthy method for the estimation of vitamine
in unknown samples.
In this connection there has recently been proposed by Mr.
Roger J. Williams a very ingenious procedure, which, if the
anticipations of its utility are realized, may prove of the greatest
assistance in the solution of the problem as to the chemical
nature of vitamine. This method is based upon the observation
that yeast requires vitamine for its growth, and the amount of
growth depends upon the quantity of vitamine present in the
culture medium. The period of the test is relatively short, and
the manipulations and apparatus are simple. A synthetic
culture medium containing asparagine, ammonium sulfate, sugar,
and salts is treated with known amounts of the vitamine solu-
tion, sterilized, and seeded with a suspension of a known weight
of yeast taken from the center of a fresh Fleischmann's yeast
cake. The amount of growth which occurs within 18 hrs.
at 30° C. is determined by filtering the yeast on a prepared
Gooch crucible, drying at 103 °, and weighing. The weight
is reported to be directly proportional to the amount of vitamine
in the solution.
LOSS OF VITAMINE ACTIVITY ON FRACTIONATION
One point upon which there is general agreement by most
investigators is that the active material is rapidly dissipated
during the several manipulations involved in the isolation pro-
cess. Each successive fractionation yields products of diminish-
ing vitamine activity. Considering the relative stability of
vitamine in its natural state, the reason for its rapid loss of
activity, when separated from most of the substances with which
it is associated in foodstuffs, is difficult to explain. An ingenious
assumption in this connection was made some years ago by Mr.
74
THE JOURXAL OF IXDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
R. R. Williams, formerly of the Bureau of Chemistry. He-
suggested that the activity is associated with the tautomeric
change which the vitamine complex, in all probability, easily
undergoes. He even thought it possible, and sought to find
the conditions under which the change from active to inactive
form, and rice versa, takes place. With such knowledge it would
be expected that the activity could be restored to vitamine con-
centrates, which had become inactive through fractionation
processes. All experiments along this line, however, were un-
successful. Although this hypothesis of Williams is very in-
teresting and suggestive, it is obviously impossible to obtain ex-
perimental support for it at this stage of our knowledge of vita-
mines.
NATURE OF VITAMINE ACTIVITY
As already mentioned, there have been a number of investi-
gators who have reported the successful isolation of vitamine
in a more or less pure crystalline state. In practically all cases,
however, the crystalline products, although in each case showing
more or less activity by physiological tests, have turned out to
be well-known compounds, such as nicotinic acid, adenine, choline,
betaine, guanine, etc., in which the vitamine function could not
be expected to reside. The activity noted in such compounds
as these can, no doubt, be best explained on the assumption
that vitamine was present in or on the crystals as an impurity.
When it is remembered that relatively minute amounts of these
crystalline substances have been found to produce physiological
results, which must have been due to the vitamine present as
a contamination, the exceedingly small amount of vitamine
necessary to produce a noticeable effect is realized. This
raises the question as to whether vitamine takes part as such in
the nutritional processes, and is converted into less complex
compounds, in the same manner that the other ingredients of
the diet are transformed," or simply acts by its presence in in-
finitesimally small amounts, in the same way that enzymes and
co-enzymes accelerate certain chemical reactions.
The view that vitamine is metabolized exactly like the other
constituents of a normal diet was early adopted by Funk, and
it is this conception which has been tacitly accepted so far.
Judging from the nervous symptoms and fatty degeneration of
the nerve cells in vitamine deficiency, Funk considered it most
probable that vitamine is necessary for the metabolism of the
nervous tissue. Thus, he states:
The lack of vitamine in the food forces the animal to get
this substance from its own tissues. The result is an enormous
loss in weight. After this available stock begins to be scarce
there is a consequent breaking down of the nervous tissue,
with the result that nervous symptoms, such as are observed
in beri beri, manifest themselves.
The conception that vitamine plays the part of an enzyme has
recently been developed in considerable detail by F. M. R.
Walsche.1 This observer considers that the reported properties
of the antineuritic vitamine suggest the probability that it is
an enzyme and is concerned directly in the hydrolysis of carbo-
hydrates.
Walsche first calls attention to the experimental evidence
that vitamine influences carbohydrate metabolism in a marked
degree. He points out that Maurer, Funk, Braddon, and Cooper
have shown a direct relationship between the amount of carbo-
hydrate ingested and the rapidity of development of polyneuritis.
Funk concludes from his experiments that increasing amounts
of foodstuffs rich in carbohydrate hasten the onset of polyneuritis,
and, consequently, that vitamine plays a more important part
in carbohydrate than in other metabolism. The evidence in
regard to the influence of vitamine on carbohydrate metabolism,
therefore, appears to be well established.
It is next pointed out by Walsche that the clinical picture of
beri beri and polyneuritis accords more with an intoxication,
due to aberrant metabolism products of carbohydrates, resulting
1 Quart. J. Med., 11 (1917-18). 320
from absence of a specific accessory factor, than with a slowly
progressive diffuse degeneration of the nervous system, resulting
from a deficiency of a nutritive constituent required for this tissue.
These observations are believed to lend weight to the view
that the action of vitamine is of the type attributable to an
enzyme. It therefore appears of interest to compare the es-
tablished properties of vitamines with those of enzymes, and
ascertain if there are any characteristic differences which would
make it improbable that the two belong to the same general
class of substances.
COMPARISON OF VITAMINES WITH ENZYMES
Considering first the source of vitamines and of enzymes, it
is to be noted that both frequently occur together. Yeast,
which is perhaps the most prolific source of vitamine, also con-
tains several enzymes, namely, glyoxalase, invertase, and others.
The castor-oil bean and many fruit juices which furnish vitamine
also contain various enzymes.
In regard to the stability of vitamines and of enzymes towards
heat, it has been found that in aqueous solutions the antineuritic
vitamine is not destroyed at the boiling point, but is destroyed
when heated to no° for 2 hrs. In the dry state, in combination
with fuller's earth, it can be heated to at least 2000 without
appreciable destruction. The antiscorbutic vitamine, on the
other hand, is known to be much less stable toward heat and
drying than the antineuritic vitamine. In the case of enzymes,
the evidence appears to be that as a rule they are destroyed by
exposure to a temperature somewhat below 100°. It is not
known, however, whether the loss of activity caused by heating
is due to destruction of the enzyme, or due to some change in
the other components of the complex colloidal system of which
the enzyme forms a part. It cannot be said, therefore, that
enzymes may not be found, or the conditions realized, under
which a temperature equal to that withstood by the antineuritic
vitamine may not prove destructive.
Since, as mentioned above, vitamines and enzymes frequently
occur in the same raw material, similar methods for their re-
moval are employed. These may involve the use of the same
solvents or other purification agents. The solubility relation?
of the two classes of substances are, therefore, quite similar.
Both vitamines and enzymes readily form adsorption com-
pounds. This would indicate that vitamine possesses the same
colloidal type of structure as is believed to be common to enzymes.
On the other hand, it has been found that the antineuritic vita-
mine dialyzes readily through parchment paper. This raises a
doubt as to the colloidal character of the antineuritic vitamine.
As pointed out by Walsche, however, there are other substances
which show all the usual characters of colloids and pass slowly
through parchment paper. The colloidal aniline dyes exhibit
all degrees of diffusibility, while in invertase and diastase we
have examples of diffusible enzymes.
The next characteristic of vitamine which may be considered
is the ease with which the activity is destroyed in alkaline solu-
tion. Considering enzymes from this standpoint it is known that
some are active only in acid and others in an alkaline medium.
The instability of both vitamines and enzymes, under particular
conditions of the solution in which they exist, is, therefore, a
common characteristic.
In regard to the failure of the attempts which have been made
to isolate enzymes and vitamines, the striking feature in both
cases is the progressive loss of activity during the application of
the analytical processes designed for their isolation. It is known
that since enzymes are colloids they carry down with them, by
adsorption, various constituents of the solutions from which
they are precipitated Consequently, they may show tests for
carbohydrates, proteins, etc., which gradually diminish as the
purification processes are improved. Simultaneously, there is
a loss of activity of the enzyme, probably due to the removal o:
bodies necessary for the full activity of the enzyme. The ex-
Jan., 1921 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
perience with vitamines is of a similar character. It is, there-
fore, apparent that the two groups of substances conduct them-
selves, in respect to fractionation procedures, in an entirely
analogous manner.
The one outstanding characteristic of an enzyme, which should
serve to dilTcrentiate it from everything with which it might be
confused, is the property of accelerating chemical reactions,
without itself being destroyed. This has been demonstrated,
to a certain extent at least, in the case of some of the well-
characterized enzymes. That it can be shown in the case
of vitamines, however, is out of the question at present, since the
only test of the activity of a vitamine is by means of a living
organism, and in such cases the recovery of the vitamine at the
conclusion of its period of action is obviously impossible.
There is this in common, however, that the apparent amount
of vitamine required for a given result is of the same order of
magnitude as required for the transformations effected by
enzymes. Thus, for instance, it has been shown that invertase
can hydrolyze 200,000 times its weight of saccharose, and rennet
can clot 400,000 times its weight of caseinogen in milk. In
the case of vitamine, fractions have been prepared of which only
a few tenths of a milligram per day are sufficient to supply the
requirements of a pigeon maintained on a vitamine-free diet.
On the basis of the above comparison it is seen that, aside
from a possibly significant degree of dialyzability, there is no
outstanding evidence that vitamines should not be classed with
the enzymes.
This viewpoint is further strengthened by the negative evi-
dence that, even in spite of the repeated efforts of able investi-
gators, the original conception that vitamine is a well-charac-
terized chemical individual capable of being isolated has never
been realized. In conclusion, therefore, the question may well
be raised as to whether our knowledge of vitamines will not be
more rapidly advanced by tentatively including them in the
class of substances designated as enzymes.
THE MECHANISM OF CATALYTIC PROCESSES1
By Hugh S. Taylor
Princeton University, Princeton, New Jersey
HETEROGENEOUS CATALYSIS
!n reviewing the general field of contact catalysis, attention
cannot but be directed to the diversity of views obtaining in
reference to the mechanism of tha process, manj' of which are
capable of direct experimental check, which, unfortunately,
in so many cases, is not applied. Sabatier1 suggests that hy-
drogenation and dehydrogenation processes occurring in con
tact with finely divided metals are to be ascribed to the capacity
of these metals to form unstable hydrides which interact with
the other components of the system to yield the reaction prod-
ucts. Thus, for the catalytic hydrogeuation of ethylene in
contact with nickel, Sabatier suggests the following scheme:
H2 + Ni; = Ni2H2
Ni2H2 + C2H, = C2H6 + Ni2
Bancroft3 suggests that it seems natural to assume that the
selective adsorption of the reaction products is the determining
factor. This conclusion, however, Bancroft shows, is not en-
tirely satisfactory in view of the known experimental behavior
of certain reactions studied. Thus, ethylene can be produced
by catalytic dehydration of alcohol by means of alumina even
in the presence of a large amount of water vapor. The beauti-
ful^studies of catalytic actions at solid surfaces recently made
by_Armstrong and Hilditch4 lead to a conclusion which is the
1 Abstract of a lecture given before the New York Section of the Amer-
ican Chemical Society, December 10, 1920.
s "La Catalyse en Chimie Organique," 2nd Edition, 1920, p. 60.
3 Presidential Address, American Electrochemical Society, April 1920.
< Proc Roy. Soc, 96 (1919), 137, 322; 97 (1920), 259, 265; 98 (1920), 27.
antithesis of the views of Sabatier. Armstrong and Hilditch
are inclined to regard the affinity of the carbon compound rather
than that of the hydrogen to the metal as of prime importance,
indeed, as the determining factor. In the hydrogenation of
unsaturated oils their experimental data lead them to the con-
clusion that the process of catalytic hydrogenation in the solid-
liquid state involves the primary formation of an unstable
complex or. "intermediate compound" between nickel and the
unsaturated compound. Dehydration reactions subsequently
studied lead them to similar conclusions in reference to primary
formation of nickel-organic compound complexes. Lewis1 as-
sumes that the mechanism of hydrogenation involves essen-
tially the dissociation of hydrogen, either adsorbed on or ab-
sorbed by the nickel, followed by collisions between the charged
nickel particles and the unsaturated molecules. He concludes
that, in the case of hydrogenation of olein and of similar sub-
stances, adsorption of the unsaturated compound on the metal
does not take place, the adsorption being restricted to metal
hydrogen components.
Many observations made in the course of experimental work
at Princeton tend to show that in the case of a variety of different
substances there occurs a definitely measurable adsorption by
catalytic agents of one or other of the reactants in a catalytic
change. In the study of the reaction kinetics of various cata-
lytic processes, indirect evidence has led to the conclusion that,
inter alia, benzene vapor is strongly adsorbed by nickel, and carbon
monoxide by nickel at temperatures as high as 150° C. Car-
bon dioxide is apparently adsorbed by iron oxide at tempera-
tures up to 250° C. Water vapor is adsorbed by various metal
catalysts. Systematic study of the magnitude of the ad-
sorption effect with a series of gases and a variety of catalytic
agents has, therefore, been undertaken. The preliminary re-
sults obtained are remarkable and serve to show the advances
in our knowledge of mechanism of catalytic change which may
come from such experimental study.
nickel — With Mr. A. W. Gauger, the adsorptions by nickel
of hydrogen, carbon monoxide, carbon dioxide, and ethylene,
using nitrogen as the reference gas have been determined in the
temperature ranges in which these gases react with one another.
The material used was reduced nickel on a porous support of
Non-Pareil Diatomite Brick, 7.5 g. of the material being em-
ployed, containing 0.75 g. of metallic nickel. The porous sup-
port used was graded between 8- and 10-mesh sieves. Table I
shows the cubic centimeters of different gases measured at 0° C.
and 760 mm. pressure which were required to fill the vessel con-
taining the nickel catalyst at 760 mm. pressure and various
temperatures.
Table I
. Temperature of Absorption Vessel, ° C. .
Gas 21 175 200 225 250 275
Nitrogen 15.04 9.8 9.4 8.8 8.5 8.1
Hydrogen 13.6 13.2 ... 12.0
Carbon dioxide 11.1 10.6 9.9 9.4
Carbon monoxide 14.05
Ethylene 14.07
If it be assumed that the adsorption of nitrogen by nickel
is negligible, the following values for the adsorption of different
gases, per gram of nickel upon the given porous support, are
readily derived.
Adsorption in Cc. (at 0° C. and 760 Mm.) per Gram Ni
Temperature ° C. 175 200 225 250
Hi 5.2 5.1 ... 4.73
COi 1.7 1.6 1.5 1.33
CO 5.66
CjH. 6.5
With ethylene, only one set of experimental measurements
has, as yet, been made. It suffices, however, to show that this
gas is more adsorbed than any of the other gases studied. With
carbon monoxide, the measurements have been limited to the
one temperature because, at lower temperatures the question
1 J. Chem. Soc, 117 (1920). 623.
76
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. r
of the formation of nickel carbonyl, Ni(CO)i, would necessarily
intrude. Above the temperature of 175° C. the measurements
are complicated by the catalytic decomposition of carbon mon-
oxide to form carbon and carbon dioxide
2CO = CO, + C.
The adsorption of carbon dioxide is noteworthy, although smaller
in magnitude than that of the other gases studied. With hy-
drogen, the initial adsorption effect is followed by a secondary
slow solubility effect which causes a slow increase in the volume
of gas required to fill the reaction vessel. This secondary change
is, however, so slow that the initial adsorption effect can readily
be measured with an accuracy of 1 per cent.
The results thus obtained with nickel may be generalized.
The gases which take part in the following reactions:
CO -(- 3Ha = CH, + HsO
C02 + 4H« = CH4 + 2H,0
C2H4 -{- H, = C^He
are all markedly adsorbed by a nickel catalyst in the temperature
range in which they react to form the stated reaction products.
copper — Similar experiments with these gases have been
performed by Mr. R. M. Burns, employing copper obtained by
reduction, at low temperatures, of copper oxide. The oxide was
produced by calcination of the nitrate in a stream of air. The
interest attaching to this study arises from the observations of
Sabatier with respect to copper as a catalytic agent. Sabatier
states that, under no conditions, can copper induce the inter-
action of carbon monoxide or carbon dioxide with hydrogen to
form methane. On the contrary, above 160° C, ethylene and
hydrogen react in contact with copper to yield the saturated
hydrocarbon, ethane. Preliminary experiments showed that
the adsorption effects with this metal were of a much lower order
of magnitude than with nickel. Consequently a larger sample
of reduced metal, 22.9 g., was used for the determinations.
The measurements of adsorption were made at 25° C, 110° C,
and 2180 C, at a pressure of 760 mm. The gases studied were
again nitrogen, carbon monoxide, carbon dioxide, hydrogen, and
ethylene. As a check on the nitrogen determination, to show
that the figures obtained with this gas represented zero adsorp-
tion, one determination was made at 25 ° C, with a specially
purified sample of helium, obtained through the courtesy of
the U. S. Bureau of Mines. Table II shows the number of cubic
centimeters of the different gases (measured at 0° C. and 760
mm. pressure) which are required to fill the reaction vessel con-
taining the reduced copper, when this is maintained at the three
stated temperatures.
Table II
Cc. Gas Required to Fill Vessel at
Gas 25° C. 110° C. 218° C.
Helium '..... 22.35 ...
Nitrogen 22.4 17.46 13 9
Hydrogen 22.4 17.6 13.9
Carbon dioxide 22.55 17.5 13.9
Carbon monoxide 23.9 18.1 13.9
Ethylene 24.1 18.1 13.9
The experiments show that only with ethylene and carbon
monoxide is there a measurable adsorption and with these gases
only at the two lower temperatures. At the temperature of
2i8° C, the volume of gas adsorbed is immeasurably small
in every case.
The experiments with copper and with nickel both show,
therefore, a greater adsorption of the unsaturated compound
than of hydrogen. It is the view of Armstrong and Hilditch
rather than that of Sabatier and Lewis which the present experi-
mental observations, therefore, tend to support, though natu-
rally a wide extension of the experimental range will be necessary
before any definite conclusions can be reached. This extension
is in progress. We are engaged on measurements of ad-
sorption with a wide variety of metals and metallic oxides under
varied conditions.
In connection with the adsorption experiments with ethylene
on copper., it is interesting to note that at the temperature at
which hydrogenation commences (i6o°) the adsorption of ethyl-
ene is already quite low. In other words, at this temperature,
the ethylene evaporates rapidly from the copper surface after
condensation has occurred. The experimental results obtained
with the gas at lower temperatures show that the copper sur-
face must be relatively free from adsorbed ethylene at the
temperature of hydrogenation. This is probably true also in
the case of the nickel experiments previously described. This
factor appears to us to be of cardinal importance in a discussion
of the mechanism of contact action. Furthermore, the fact
that, as far as adsorption by copper is concerned, carbon monoxide
behaves like ethylene, whereas hydrogenation of carbon mon-
oxide in contact with copper cannot be achieved, shows that
further insight into the several factors prevailing is still needed .
We propose to obtain this by extending our studies on adsorp-
tion by various metallic catalysts .which either promote or are
inert in the hydrogenation process. Thus, in contact with co-
balt, carbon monoxide and hydrogen yield methane. With
iron, no methane is obtained.1
Since carbon monoxide and hydrogen do not interact in con-
tact with reduced copper it is possible to study the adsorption
of these gases from mixtures of the same. Similar studies can
be carried out with mixtures of ethylene and hydrogen at tem-
peratures below those at which these gases interact. In a
preliminary manner we have studied the adsorption of various
mixtures of hydrogen and carbon monoxide and hydrogen and
ethylene at 25 ° C. The results obtained are very remarkable
and promise further insight into the catalytic process. In
Table III are given the adsorptions in cubic centimeters of gas
absorbed by 22.9 g. of reduced copper with various mixtures
of the two pairs of gases. In the last column are given the cal-
culated values for adsorption, if the amounts adsorbed were
in direct proportion to the partial pressures of the gases present.
Table: III
Cc Gas Calculated Adsorption
(at 0°, 760 Mm.) if Proportional
Absorbed at 25° C. to Partial Pressures
Gas Mixture and 760 Mm. of Gases
0% Hi, 100% CO 1.5
50% H:, 50% CO 1.3 0.75
84.5% H2, 15.5% CO 0.9 0.23
100% Hs 0.0
0% Hi. 100% C3H4 1.7
53% Hs, 47% CjH. 1.2 0.8
100% Hi 0.0
It is thus apparent that carbon monoxide and ethylene are
much more markedly adsorbed at lower pressures than at higher
pressures, the adsorption tending to become independent of the
pressure as this increases.
THE KINETICS OF CATALYTIC ACTIONS
The abnormal variation of adsorption with pressure consti-
tutes a factor of considerable importance in regard to the mech-
anism of the catalytic process. If the catalytic reaction occur*
in the surface layer it is apparent that the pressure-adsorption
ratio determines the concentration of the reactants in the active
layer. For example, in the reaction
C2H4 "r H2 = C2H6
the rate of formation of ethane in the gas phase is
Ri = fc(/>C*H.)(fc&).
-where pc*Ri and pHi are the partial pressures of the inter-
acting gases. Similarly at the surface of the copper, the rate
of reaction is
R2 = fe(Cc2H.)(CH,1.
where Cc2H( and Ch* are the concentrations of the gases at the
surface. Now, if the experimental conditions were so chosen
that the concentration of ethylene in the surface layer was inde-
1 Sabatier, hoc. cil.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
77
pendent of the prevailing partial pressure of the gas, i. e ,
CCiH, = HpC2Kt)° = k,
the reaction in the surface layer would become
R2 = fc.fe.(CHs).
So, if the hydrogen concentration were governed by Henry's
law, the reaction would be bimolecular in the gas phase and ap-
parently monomolecular in the surface layer.
The same considerations might be extended to the question
of the equilibrium constant of the given reaction. In the case
cited, with the same assumptions as to the distribution ratio
between gas and surface layer, the equilibrium constant Kg
in the gas phase would be
K = ki(pc,H,)(pH,)
kziPdHe)
In the surface layer, however, if the hydrogen and ethane obeyed
Henry's law, but the ethylene concentration was independent
of the partial pressure of ethylene, the equilibrium constant,
Kj, would be
_ fe.fc.(CH.)
' WCCH.) '
It is apparent, therefore, that the position of equilibrium in
the surface layer could be markedly different from the true
equilibrium in the gas reaction. It cannot, however, be too
strongly emphasized that this does not mean that a catalyst
can shift the equilibrium of the gas reaction. The equilibrium
in the gas phase remains identically the same as it would be if
achieved thermally without a catalyst. An analogous case,
with two solutions, is that studied by Kuriloff,1 who investi-
gated the equilibrium between /3-naphthol and picric acid in
water and benzene solutions, in presence of solid picrate. The
product of the millimolar concentrations of free naphthol and
free undissociated picric acid varied widely in the two solvents,
being 2.89 in water and 7550 in benzene, in agreement with the
deductions from distribution experiments of the individual
substances. The presence of a benzene layer adjacent to the
aqueous layer, however, did not in any way disturb the equilib-
rium in the aqueous layer.
Table IV
1 (Hours)
*(SOi)
0.5
12
1.0
20
1.5
27
2.0
32
2.5
36
3.0
40
3.5
43.5
4.0
46.5
5.0
52
6.0
57
7.0
62
8.0
67
9.0
72
10. 0
76
11.0
80
■ 12.0
84
As a consequence of' these considerations it follows that the
study of the kinetics of catalytic reactions may give reaction
equations totally different from those to be expected from the
stoichiometric equation for the gas reaction. This is well
known from the experimental work of Fink on the mechanism
of the formation of sulfur trioxide from sulfur dioxide and oxygen,
of Bodenstein and his co-workers on carbon monoxide and oxygen,
and from the recent studies of Armstrong and Hilditch in liquid
media. Furthermore, since, as the experiments cited previously
show, the distribution of gas between the reaction space and
catalyst surface is different at different partial pressures, it fol-
lows that a given equation for the reaction kinetics, while valid
over one pressure range, may be invalid over another pressure
range. This is clearly shown in many of the kinetic studies
1 Z. physik. Chem., 26 (1898), 419.
quoted. Fink's results on sulfur trioxide formation show no
agreement with a termolecular reaction equation in the early
stages of an experiment. Towards the completion of the pro-
cess, however, an excellent termolecular constant, k3, is obtained
as Table IV shows.
On the interpretation given in the preceding paragraphs the
distribution of sulfur dioxide and oxygen between the gas phase
and the contact material must in the later stages of the reaction
follow Henry's law.
Bodenstein and Ohlmer found that the reaction between
oxygen and carbon monoxide in contact with quartz glass takes
place at a rate proportional to the pressure of oxygen and in-
versely proportional to the pressure of carbon monoxide. In
contact with crystalline quartz, however, the reaction followed
the ordinary stoichiometric equation, a result which should
have attracted a much greater attention in the discussion of
catalysis than it has yet done. On the interpretation here given,
this diversity of reaction mechanism, in the same reaction,
with the two catalysts, is to be ascribed to the different dis-
tribution ratios between the gas phase and the surface layer on
the contact mass. An experimental test of such a viewpoint
could be carried out.
HOMOGENEOUS CATALYSIS
For catalytic reactions in homogeneous systems the inter-
mediate compound theory appears to be generally applicable
For most such processes a probable cycle of successive reactions
can be postulated. In many cases the intermediate compounds
have been isolated. In other cases, the indirect evidence lead-
ing to such a conclusion is being steadily brought forward. For
example, Jones and Lewis' give evidence for the formation of an
intermediate sucrose-hydrogen-ion complex in the sugar inver-
sion process. In ester hydrolysis the systematic researches of
Kendall and his colleagues2 have established the existence of
binary and ternary compounds between ester, catalyzing acid,
an ' water. The tendency towards compound formation is the
more marked, the greater the chemical contrast between the
basic nature of the ester and the acidity of the catalytic agent
The concordance of this conclusion with the observation that
the catalytic activity in ester hydrolysis is greatest with the
strong acids and diminishes with decreasing strength of acid
forms a striking piece of evidence in favor of the intermediate
compound theory in such systems.
Development of the radiation theory of chemical action
(Trautz, Lewis, Perrin) has led to the supposition that the neces-
sary energy of reaction is supplied by suitable infra-red radia-
tion. In the beginning, the attempt was made simply to as-
sociate the critical energy increment with the heat of reaction
and to show that such relationships were plausible in view of the
infra-red absorption bands shown by the reacting substances.
Recently, Rideal and Hawkins3 have attempted to show that
infra-red radiations actually accelerate the velocity of hydrolysis
of methyl acetate. A pronounced positive result is claimed.
The conclusion, however, can be accepted only with reserve,
for the experimental conditions, as far as they may be de-
duced from the publication, were not ideal. Indeed they were
such that, if the positive effect attained is real, the magnitude
of the effect of the infra-red radiations must be enormous. The
experiments were carried out with 100 cc. of an aqueous solution
containing catalyzing acid and ester. The radiation was intro-
duced into the system from above. Owing to the opacity of
water to infra-red radiation it is therefore evident that only a
film of solution in the surface layer was being irradiated. Since
the stirring was only occasional, it is apparent that hy fir the
greater bulk of the solution was not acted upon by the infra-
' J. Chem. Soc . 117 (1920), 1120.
« J. Am. Chem Soc, 1914, el seq.
> J. Chem Sot., 117 (1920). 1288.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. n, No. i
red rays. These could be distributed through the solution only
by diffusion of the activated hydrogen ions or hydrogen-ion-
i/ster complexes from the surface into the interior.
The question, however, of the possible activity of infra-red
experiments should be undertaken. If such be done the choice
of a reaction system through which the radiation might readily
penetrate would facilitate the attainment of decisive experimental
test. We hope to take such problems in hand at an early
rays is so important that duplication and amplification of such date.
INDUSTRIAL AND AGRICULTURAL CHLMI5TRY IN THL BRITISH
WL5T INDILS. WITH SOML ACCOUNT OF THL WORK OF 5IR
FRANCIS WATTS, IMPLRIALCOMMISSIONLR OF AGRICULTURL
By C. A. Browne
X. V Sugar Trade Laboratory, 80 South St., Nsw V.
Received October 5, 1920
The casual traveler, who makes his first voyage among the
West Indian Islands and views from his steamer the crumbling
walls of old fortresses, or the remains of stone mansions, acquires
at the outset the feeling of a departed civilization. This first
impression is intensified by the ruined walls and towers of ancient
muscovado sugar works, which, according to the lines of Grainger,
the poet of St. Kitts, were once lit up at night by "far-seen
flames bursting through many a chimney." It is only when
the vessel steams past these scenes of desolation into the harbor
of Basseterre, the former home of this poet, and the smoking
stacks of a modern sugar factory come into view that the im-
pressions of decadent or vanished industries are dispelled.
The present paper is an effort to tell briefly the story of this
change from an old to a new order of things, in which transition
the efforts of a distinguished member of the American Chem-
ical Society have played a prominent part.
With the abolition of slavery in the British West Indies in
1834, the old industrial system of these islands came to an end.
The production of sugar, which had always been the chief source
of wealth, began to decline, partly from lack of labor and partly
from unequal competition with the more scientifically conducted
beet-sugar industry of Europe, which marked its phenomenal
rise from the date of the abolition of slave labor in the colonies.
The inequality of this conflict was later enhanced by the favoring
export bounties which beet sugar received, and had it not been
for the high prices of sugar, which existed for 20 years
after the outbreak of the American Civil War, the declining
sugar industry of the West Indies would have completely dis-
appeared.
The over-stimulation of the beet-sugar industry by bounties
and premiums soon had, however, its inevitable effect, and
between 1882 and 1892 the price of muscovado fell from 7.3
cents to 2.8 cents per pound. The industrial condition of the
British islands was becoming hopeless, and appeals were made
for assistance to the mother country, which for the 50 years
following the abolition of slavery had shown a strange indifference
to its West Indian possessions. This neglect had in fact become
so marked that many planters believed their only hope to consist
in political union with the United States. It was only with the
growing development of the Panama Canal enterprise in the
late eighties and the dawning sense of the future strategic and
economic importance of the island approaches to this gateway
of the Pacific that Great Britain began to take a renewed in-
terest in her tropical colonies. From that time until the present,
increasing efforts have been made to improve the industrial,
economic, and educational life of the British West Inches.
Botanic gardens, experiment stations, and other scientific
institutions were established, among the earliest of these being
the government laboratory in the island of Antigua, which
began its work on Jan. 1, 1889, and of which Dr. (now Sir)
Francis Watts, a graduate of Mason College, Birmingham,
assumed charge as analytical chemist.
IMPROVEMENTS IX SUGAR MANUFACTURE
One of the first investigations which Dr. Watts instituted on
beginning his new duties was a thorough examination of the
field and factory methods of the sugar industry. His chemical
training convinced him that if the cane sugar of the West Indies
had to compete with the more scientifically manufactured beet
sugar of Europe, the wasteful antiquated processes of the little
muscovado factories must disappear.
In a little work, entitled a "Manual for Sugar Growers,"
and in various reports, Dr. Watts opened the eyes of the West
Indian planters to the enormous losses which their small factory
system involved, and as a remedy suggested the erection of large
scientifically managed central factories. The idea was favorably
received but opinions were divided as to whether such factories
should be under government or private control. After much
discussion a working scheme was evolved, whereby a group of
British capitalists negotiated contracts with certain estate owners
in Antigua under which the latter undertook to supply, during
a period of 15 years, the sugar canes grown on certain stipulated
areas at a price based on the current market price of sugar,
coupled with a share in the profits of the factory and, ulti-
mately, a share in the ownership of the factory itself to the extent
of one-half. The capitalists formed a company with a capital
of some $200,000, including a sum of $72,000, subscribed by
the government. With this a small central sugar factory
HI
jJHj
ifePr
laliiP
Old Muscovado Sugar Factory, British West Indies
capable of making about 3000 tons of sugar in a season, was
erected at Gunthorpes, Antigua. The success of the new enter-
prise was immediate, and the Antigua factory has now grown
from a capacity of 3000 to 10,000 tons of sugar per season.
In 1919, at the end of the 15 years' agreement, the government
cancelled its $72,000 subscription, its own income from the
enterprise in the form of excess profits and exports taxes having
exceeded $300,000. The contracting planters received during
this time an average of 20 per cent annually on their original
investment, and at the end of the 15 years had turned over to
j£
1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
79
them shares representing $250,000 and approximately $90,000
to their credit on the company's books.
The belief that the Antigua central factory would be a pioneer
object lesson for sugar planters in other islands was so well
vindicated that a second cooperative factory was soon estab-
lished at Basseterre in the island of St. Kitts, and others in
Barbados, Trinidad, and Jamaica. The success of these central
factories has naturally had a most favorable influence upon the
welfare of the islands, the laborers receiving the benefit in in-
creased wages and the small farmers in the increased price for
their canes. All this prosperity has resulted from the simple
fact that with the economies of the chemically controlled central
system only about 9 tons of sugar cane are needed to make a
ton of sugar, while with the primitive muscovado process 14 to
15 tons of cane were required.
The results of the Antigua factory for the years of its operation
are summarized in Table I.
Table I — Results of the Antigua Sugar Factory
1905-7 1908-10 1911-13 1914-16 1917-19
3 Yrs. 3 Yrs. 3 Yrs. 3 Yrs. 3 Yrs.
Average Average Average Average Average
Cane ground, tons 27,106 42,888 61,612 92,302 85,690
Sugar made, tons 2,737 4,693 6.349 9.970 9,586
Sucrose in cane, per cent 14.17 14.37 13.74 12.67 12.79
Sucrose in bagasse, per
cent 7.33 6.07 4.61 3.22 2.63
Purity of juice, per cent 87.60 85.38 83.70 83.90 83.83
Recovery of sucrose, per
cent 68.43 73.10 72.18 82.06 84.15
Yield of sugar, per cent.. 10.03 10.93 10.32 10.78 11.20
Price of sugar, per ton... $49.68 556.42 $53.66 $69.60 $103.68
The results show that while there has been a marked increase
from year to year in factory efficiency, as shown by the rising
recovery of sucrose and the diminishing loss of sugar in bagasse,
this gain has been offset by a progressive decrease in the sucrose
content and purity of juice in the cane. The latter circumstance
has given rise to the fear that the cane of Antigua might be
undergoing a degeneration like that of the Bourbon cane in
the West Indies about 1890 and of the Cheribon cane in Ar-
gentina in 1916. The probabilities, however, are that the
diminishing sucrose content of the sugar cane in Antigua is due
to certain defects of the central system, especially in times of
shortage and ascending prices, whereby cane cutters and plant-
ers, from being paid by quantity instead of by quality, send to
the factory a large amount of cane that is unripe, diseased,
trashy, or otherwise unfit for milling. The spoiling of cane by
fermentation, as a result of delays between cutting and milling,
is also no doubt responsible for much of the trouble,1 a supposi-
tion which is confirmed by the fact that the fiber content of the
cane at the time of grinding has increased from its original value
of 15 per cent in 1905 to 17 per cent. The excess of fiber in
the sugar cane of Antigua, while insuring an extra sufficiency of
bagasse for fuel, has its objection in that the difficulties of milling
are vastly increased. This factor in an island of insufficient
rainfall and inadequate water supply, such as Antigua, where
maceration must be curtailed, necessarily impairs the recovery.
The central factories of Antigua and St. Kitts were visited
by the writer during the campaign of 1919. Both establish-
ments are thoroughly modern in their equipment and the con-
trast between them and the few remaining muscovado factories,
that were still in operation, was most striking.
CANE SIRUP
Closely connected with the sugar industry of the British
West Indies is the manufacture of cane sirup or, as it is locally
termed, fancy molasses. The process is generally carried out
in the old muscovado factories, the primitive equipment of
which is well adapted to the making of sirups. The steps of
manufacture are in fact very similar to the operations of making
1 The deterioration in quality of cane supplied to the factory has also
been noted in St. Kitts and other West Indian islands. For a full discussion
of the question see papers by Sir Francis Watts in the West Indian Bulletin,
16, 96, and 17, 183; also the paper by L. I. Henzell in the Louisiana Planter,
62 (1919). 395.
muscovado, the only difference being that precautions are taken
to invert a part of the sucrose in order to prevent its crystalliza-
tion in the container. The process, as the writer saw it carried
out in Barbados, is briefly as follows:
The canes are crushed by means of wind power between three
vertical rollers, the juice from the mill flowing by gravity into
a clarifying tank where it is heated with a little milk of lime, in-
sufficient to neutralize the natural acidity. The limed juice
after heating is allowed to settle, and the clarified liquid drawn
off into a train of copper evaporating kettles, called tayches,
heated by burning sun-dried bagasse. In the first evaporator
the juice is treated with a bucket of cane juice that has under-
gone an acid fermentation, in order to invert a part of the sucrose.
The boiling liquid is skimmed to remove impurities and during
concentration is ladled from tayche to tayche until it finally
reaches a density of about 36° Be. hot, when it is run into a
cooler. The product when cold has a density of 42 ° Be., is
of a clear wine color, and has a most agreeable flavor.
The composition of several grades of "Fancy Molasses"
according to analyses made in the Antigua laboratory by Dr.
H. A. Tempany1 is as follows:
Table II — Composition of "Fancy Molasses"
I II III IV V
Water 22.4 19.7 19.8 27.1 21.9
Sucrose 46.3 42.1 43.0 44.2 51.0
Reducing sugars 27.3 32.8 30.7 24.4 20.0
Ash 1.3 1.9 1.5 3.3 1.8
Non-sugars 2.7 3.5 5.0 1.0 5.3
Total 100.00 100.00 100.00 100.00 100.00
Direct polarization. . . 39.9 35.0 35.1 36.2 47.5
Degrees Be 41.5 41.2 39.0 41.0
In Sample IV the evaporation was not carried to the proper
degree, and in Sample V the inversion was not sufficient to
prevent crystallization. A sirup of the so-called "two forties"
standard (that is, having a direct polarization of 40 and a density
of 40° Be.) will keep without crystallization, and this is the general
aim of the manufacturer.
It is unfortunate that so little of the pure cane sirup manu-
factured in the West Indies finds its way directly to the table.
A large part of it is used by blenders for mixing with low-grade
molasses, a good product being thus adulterated to improve
an inferior one. It is the opinion of many West Indian pro-
ducers that the only effective means of getting their sirup to
the consumer in a pure recognizable form is to can the product
at the factory in sealed tins, upon which the name of the brand
is stamped in raised letters.
The activities of the government laboratory in Antigua have
been directed to the improvement of other industries besides
those of sugar and sirup. The great dissimilarity between the
different West Indian islands in soil, rainfall, and other climatic
conditions has necessitated a careful study of the adaptability
of each island to special crops and industries. The precarious
condition of sugar manufacture in the islands, where the central
system is not feasible, has also led to the encouragement of other
agricultural industries. Of these we can mention only cacao,
citric acid, essential oils, and rubber.
CACAO
Next to sugar the most important agricultural enterprise
of the British West Indies is the growing of cacao.
The cacao tree becomes productive when about 5 years of
age and, if in a healthy condition, will continue to bear from 40
to 50 years. Isolated trees may attain a height of 30 to 40
ft., although in cultivated orchards the maximum is not allowed
usually to exceed 15 to 20 ft. The fruit consists of an elongated
pod, containing from 20 to 50 or more beans or seeds, embedded
in a pink colored pulp. When ripe the seeds with the adhering
pulp are removed from the fruit and, after undergoing a process
of curing or fermenting, are cleaned, dried, and packed for the
market.
1 West Indian Bulletin, 13, S24
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol.
No.
Courtesy of the
Vacuum Pans and Multiple Effects, Gunthorpes Sugar Factorv, Antigua
Iii the process of curing, as observed by the writer in Do-
minica, the pulp-covered seeds are placed under cover in boxes,
where they are turned over once or twice a day. The tempera-
ture of the mass begins to rise and in a few days attains a maxi-
mum of about 45° C. The heating of the beans is the result
of a fermentation of the adhering mucilage, the sour liquid or
sweatings," which drain from the mass, being allowed to escape.
Kmployment of this waste for vinegar making and other pur-
poses has been proposed, but so far no successful method of
utilization has been devised. After fermenting, which lasts
from 3 to 7 days, the seeds are dried in the sun on large trays
which can be wheeled on tracks under shelter in case of rain.
As a result of the fermenting process the cacao seeds are not
only freed from pulp, but a number of important chemical
changes take place which improve the character of the product.
The beans take on a brown-mahogany color, agreeable aromatic
odors and flavors are developed, and the astringent tannin sub-
stances, which give the uncured beans a bitter taste, are modified
or removed. The subsequent drying in the sun appears to
promote the changes begun in the curing house, an effect which
artificial drying by machine does not seem to accomplish. Arti-
ficial drying is necessary, however, in rainy localities in order
to prevent the beans from becoming moldy and mildewed.
The product must be dried slowly at not too high a temperature;
fans must also be used to insure circulation of air. Artificial
drying1 is most successful when the conditions of sun drying
are imitated as closely as possible.
The chemistry of cacao curing and the conditions of obtaining
the most desirable aroma and flavor are at present very im-
perfectly understood. As Knapp2 has recently pointed out,
an important and most attractive field of chemical research here
awaits investigation.
Experiments to determine the chemical conditions of soil
necessary for securing the most favorable yields of cacao were
instituted by Dr. Watts, in association with the officers of the
Agricultural Departments, in Dominica, in 1901, and the results
of this work, which have been continued for nearly 20 years,
throw a great deal of light upon the fertilizer requirements of
this particular crop. These experiments, as summarized by
Tempany,8 show that by far the best yields under Dominican
conditions are obtained from soils which have been mulched
with a nitrogenous dressing of legumes, the decomposition of
the organic matter thus supplied rendering available the natural
reserves of potash and phosphoric acid already existing in the
soil. From 3 to 5 years are required for cacao trees to acquire
1 G. Whitfield Smith, "Artificial Drying of Cacao," West Indian Bul-
letin, 2, 171.
s "Application of Science to Cacao Production." /. Sac. Chem. Ind ,
37 (1918), 468.
3 "A Study of the Results of the Manurial Experiments with Cacao
Conducted at the Botanic Station, Dominica." West Indian Bulletin, 14, 81.
Jan., 1921
TEE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
%i
the maximum productivity occasioned by any particular method
of manurial treatment.
CITRIC ACID
The citric acid produced in the British West Indies is derived
entirely from limes, the industry being confined mostly to the
islands of Dominica, Montserrat, and St. Lucia. Dominica
leads in lime production, and the exportation of lime juice,
calcium citrate, and the essential oil of limes from this island is
summarized in Table III, which is taken from statistics supplied
by Dr. Watts.1
Table III — Average Annual Exportation of Lime Products from
i-Year Period
1895-1899
1900-1904
1905-1909
1910-1914
1915-1919
127.960
249,849
208,26.5
348,108
600,720
Dominica
Concentrated
58,486
90,295
124,643
148,571
154.185
1816
4995
2718
Oil of
Limes
Gal.
2707
3983
4761
6166
6343
Ordinarily one barrel of approximately 1200 limes yields 8
gal. of raw juice (containing about 12 oz. of citric acid per
gallon) and 1 gal. of concentrated juice (containing about 6 lbs.
of citric acid per gal.). One gallon of raw lime juice yields
approximately 1 lb. of commercial calcium citrate.
In the ordinary crude process of concentration, the expressed
lime juice, after straining to remove floating impurities, is
first heated in a copper still to recover the essential oil. The
juice, after settling, is boiled down over an open fire in a train
of copper tayches, similar to those employed in the manufac-
ture of sirup or muscovado. The course of the lime juice is,
however, opposite to that followed in concentrating cane juice,
the strike being taken from the kettle furthest from the fire,
as greater losses from decomposition of citric acid occur when
the final concentration is made directly over the flame. The
degree of economical concentration is from about 9 volumes to
1 , the loss of acid becoming considerable if a higher concentration
is attempted. The final product is a thick black liquid, which
after cooling is run into 54-gal. casks for shipment. The loss of
citric acid by open-fire concentration varies from 6 to 16 per
cent.
In order to reduce the loss from destruction of citric acid, an
improvement has been made by concentrating the lime juice
in jacketed steam-heated pans. The loss of citric acid by this
method is said to be reduced to less than 3 per cent. In some
localities use is also made of wooden vats heated by steam coils.
Metal coils of tinned copper or of block tin are recommended
as the most suitable, as they are less subject to attack by the
hot concentrated acid. It has also been found that the use of
granite rollers, in place of iron, for crushing the limes, gives a
brighter, purer juice.
The objections to concentrated lime juice, due to destruction
of acid, expense for casks, leakage, freight, etc., induced Dr.
Watts2 in 1902 to discuss the manufacture of citrate of calcium.
After considerable experimenting he published a process for
manufacturing citrate from lime juice. As a result of this work,
the manufacture and exportation of citrate of calcium was
started in Dominica in 1906.
In the manufacture of citrate of calcium, as observed in
Dominica by the writer, the juice is removed from the crushed
limes by powerful presses. The expressed juice is then heated
in a still to recover the essential oil, the latter being collected
from the distillate in a Florentine receiver. After removing
the volatile oil, the hot juice is discharged into a settling tank
to deposit albumin, pectin, and other impurities. The clear
liquid, together with that obtained from the filtered settlings,
is neutralized with chalk and heated nearly to boiling, which
1 "The Development of Dominica," West Indian Bulletin, 16, 198.
""Citrate of Lime and Concentrated Lime Juice," Ibid., 2, 308; 7,
331; 9, 193.
causes the citrate of calcium to become crystalline and to settle
quickly. The clear, yellow, mother liquor is drawn off; the
precipitated citrate is washed several times in hot water, and
then pressed or separated in a centrifugal, after which it is
dried in a current of air between 150° and 2000 F. The moisture
content of the citrate should be reduced below 10 per cent, as
otherwise there is danger of destructive fermentation. The
commercial citrate of calcium thus prepared contains about
65 per cent citric acid. The losses of citric acid by this process
are reduced to about 2 per cent. The expense for chalk and the
cost of drying nullify, however, certain advantages which the
citrate industry has over concentrated lime juice, and large
quantities of the latter still continue to be manufactured.
The lime juice and calcium citrate manufactured in the West
Indies are exported to the United States and Great Britain,
where they are used for manufacturing citric acid for calico
printing, for making beverages and medicinal preparations,
and for various other purposes.
ESSENTIAL OILS
ESSENCE OF LIMES — The principal essential oil manufactured
in the British West Indies is essence of limes, which is prepared
in two forms, the attar of limes or hand-pressed oil, and the
distilled oil, which is a by-product in the manufacture of con-
centrated lime juice or calcium citrate. The attar of limes,
which is the more fragrant and valuable, is removed from the
fruit by an implement called from its French name an ecuelle
(meaning porringer). The latter consists of a shallow copper
dish with blunt projections on the inner surface and with a
hollow receptacle in the handle at the bottom. The limes are
rapidly rotated by hand across the projections, the essential
oil escaping from the ruptured cells of the skin and running down
into the receptacle. An expert native woman can extract over
30 oz. of oil a day by this process. The oil, after pouring from
the receptacle of the ecuelle, is separated from the underlying
watery fluid and filtered to remove cellular matter and other
impurities. A barrel of limes yields from 3 to 5 oz. of attar by
the ecuelle process, while the juice from a barrel of limes will
yield from 4 to 6 oz. of the distilled oil.
Analyses of West Indian hand-pressed and distilled oils,
made in the Antigua laboratory by Tempany and Greenhalgh,'
showed the following results;
Table IV — Properties of West Indian Lime Oils
Hand-Pressed Oil Distilled Oil
(Antigua, Montserrat, Dominica) (Dominica)
Specific gravity, 30° C . 0. 8659-0. 88593 0.854O-O.8858
Angular rotation, 31°, 100
mm. tube +31.38°- +33.43° +33.09°- +34.89°
Refractive index at 32° C. 1.4789- 1.4836 1.4702- 1.4713
Citral, per cent 2.2- 6.6 1.2- 2.0
Acid number 1.35- 2.8 0.76- 1.3
The distilled oil is distinguished chemically from the hand-
pressed oil by its lower percentage of citral, this aldehyde being
partially destroyed during the boiling of the acid lime juice.
bay oil — -The distillation of bay oil from the leaves of the
West Indian bay tree (Pimento, acris) is an industry of some
importance in several of the West Indian islands. One of the
earliest studies upon the production and chemical composition
of bay oil was made by Watts and Tempany2 in the Antigua
laboratory in 1910. Later experiments have been conducted
in the island of Montserrat to determine whether it might not
be more profitable to obtain bay oil from carefully selected
and cultivated stock rather than from the wild native trees
scattered through the woods. The results by Tempany and
Robson3 in Table V show, in fact, a wide difference in the yield
and character of the oil from different trees.
1 "Notes on Expressed and Distilled West Indian Lime Oils," West
Indian Bulletin, 12, 498.
= West Indian Bulletin, 9, 271.
» "Bay Oil and the Cultivation of the Bay Tree as a Crop Plant."
Ibid., 16, 176.
82
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
Table V — Yields and Properties of Bay Oil from Different Trees
Yield of Oil
per 1 00 Lbs.
Green Leaves
Fluid Ounces
12.6
6.2
18.4
17.3
24.7
19.2
Specific
Gravity
0.9822 at 29.5°
1.0051 at 30°
0.9610 at 29.5°
0.9850 at 29.5°
0.9890 at 29.5°
0.9814 at 29°
Phenol
Content
Per cent
Rotation in
100 Mm. Tube
— 1.60 at 28°
— 1.35 at 29°
—2.05 at 29°
— 1.49 at 28°
Refrac-
tive
Index
1.5155
1.5187
1.5121
1.5163
1.5161
1.5152
According to these results the selection of seed for planting
purposes, on the basis of yield and quality of oil, has a prom-
ising outlook.
Owing to the complex composition of bay oil the haphazard
methods of distillation practiced by the natives may lead to
products of widely different character. The first fraction ob-
tained by steam distillation of the leaves consists mostly of the
lighter more volatile constituents, myrcene and phellandrene,
which float upon the waste water in the receiver. As distilla-
tion proceeds, mixtures of oils come over that have the same
density as water, and from which unaided they separate with
difficulty. The later fractions consist mostly of eugenol, with
small amounts of chavicol and other phenols, which, being
heavier than water, settle to the bottom of the receiver. The
lighter oils in rising and the heavier oils in sinking dissolve and
carry with them the portions in aqueous suspension. The
mixture of the surface and bottom fractions, when distillation
is complete, constitutes the normal bay oil of commerce. Should
the receiver be changed at the wrong time, the separation of the
oil suspended in the waste water may not be perfect. The
losses from this cause and from incomplete distillation not only
diminish the yield but give rise to products of abnormal com-
position.
■ JM
.". -
.}
jELyjfe '^Vrr '"'"7 5fcffiV«r
■^Mm
N^Mi^K^M^
Headquarters of Imperial Department of Agriculture,
Barbados, British West Indies
Experiments conducted by Dr. Tempany in the Antigua
laboratory upon the changes in bay oil during storage show that
the phenol content remains unchanged but that the specific
gravity tends to rise considerably. The latter fact is explained
by the polymerization of the myrcene, a reaction that proceeds
more rapidly in the air. For this reason it is important that
vessels used for containing bay oil should be tightly closed.
thymol — At the time of the writer's visit to the Antigua
laboratory in 1919, considerable attention was being given by
the acting government chemist, A. E. Collens, to the possibility
of producing thymol1 from horse mint {Monarda punctata) and
ajowan seed (Carum copticum). Air-dried ajowan seed grown
in Montserrat gave on distillation a yield of 3 per cent of an
oil, which yielded a recovery of 43.5 per cent thymol crystals.
1 "Notes on Thymol Content of Horse Mint and Ajowan Seed,"
West Indian Bulletin, 17, 50.
The calculated yield per acre was about 35 lbs. of ajowan oil,
which, on a basis of 40 per cent recovery, would indicate a yield
of 14 lbs. of thymol per acre. This at present prices of the drug
was considered profitable.
The field and laboratory researches of the Imperial Depart-
ment of Agriculture all indicate that the essential oil industry
in the British West Indies has a most promising future.1
While the exportation of rubber from the British West Indies
has not attained a leading economic importance, a large amount
of investigation has been conducted by the Imperial Department
of Agriculture concerning the adaptability of the various rubber-
producing trees to the climatic conditions of the different islands.
In localities which have an evenly distributed rainfall of over
75 in. per year and a minimum temperature of not less than
65 ° F., such as obtain in parts of Trinidad, Dominica, and
Tobago, the Para rubber tree (Hevea brasiliensis) thrives well,
giving on properly cultivated plantations an average yield of
200 lbs. of rubber per acre. The Castilloa rubber tree grows
better in districts with a moderate rainfall, but the yield of
rubber per acre is much less than with Hevea. With the latter
tree there is a steady flow of latex nearly all the year, while with
Castilloa there is but little wound response and the trees must
be tapped at frequent intervals. The problems of tapping the
Castilloa and dealing with its latex give difficulty and have not
been perfectly solved.
Probably over three-fourths of the plantation rubber made
in the British West Indies is coagulated from the latex by means
of acetic acid ; lime juice is also extensively employed. According
to Collens,2 the cheapest and most efficient coagulating agent is
a 5 per cent solution of sulfuric acid, in the proportion of 10
drops to 100 cc. of latex. The rubber coagulated by this means
was found to be of excellent quality and showed no signs of
deterioration.
In the process employed on plantations, the clotted cream,
which rises to the surface of the coagulated latex, is gently
washed, pressed, and then allowed to dry for a day. The "bis-
cuits" of rubber thus prepared are then smoked for 3 or 4 days
until they become transparent, during which interval they take
on an amber color and acquire a characteristic smoky smell.
The chief obstacle to the development of plantation rubber
in the British West Indies is the scarcity of cheap labor; for
this reason it is doubtful if the industry there will ever achieve
the same degree of success as it has gained in Ceylon and the
Malay States.
Limitations of space prevent the description of other tropical
industries such as those of the starches, vegetable oils, tanning
materials, dyewoods, and copra, in which there is much of
chemical interest both general and special. The extensive
chemical investigations of the Antigua laboratory upon water
supplies, soils, mineral deposits, and matters pertaining to the
public health, as well as the important researches of Dr. Watts
and Dr. Tempany in improving methods of analysis, must also
be passed over in order that a few words may be said about the
development and future of scientific research in the British West
Indies.
THE WORK OF SIR FRANCIS WATTS
The early work of the Antigua laboratory, when Dr. Watts
assumed charge in 1889, was begun in great isolation and under
enormous difficulties. The laboratory appliances were meager,
there was no gas, the library consisted of only a few general
works and there was no consulting staff of scientific co-workers;
yet this lack of equipment, denoting as it did the complete
> For the almost unlimited possibilities in this field see article by J. H.
Hart, "Preparation of Essential Oils in the West Indies," West Indian
Bulletin, 3, 171.
« "Rubber Experiments in Trinidad and Tobago," Ibid., IS, 219.
Jan., 1921 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
83
absence of any predetermined governmental policies, left the
laboratory free to develop along natural lines and to take up
the industrial and agricultural problems of most immediate
and pressing importance. The great benefit of the laboratory
was quickly felt and the scope of its work was widened when,
Sir Francis Watts, K.C.M.G., D.Sc.
Imperial Commissioner of Agriculture for the West Indies
with the establishment of the Imperial Department of Agri-
culture for the West Indies in 1898, the local Antigua laboratory
became a federal institution, with its field enlarged to comprise
St. Kitts, Nevis, Montserrat, and the Virgin Islands. Imme-
diately preceding this. Dr. Watts occupied for about a year the
position of chemist to the government of Jamaica, but re-
linquished this post after the creation of the Imperial Depart-
ment, to accept in 1899 the appointment of government chemist
and superintendent of agriculture for the Leeward Islands.
He retained this position until January 1909, when he was ap-
pointed to his present office of Imperial Commissioner of Agri-
culture for the West Indies.
From the beginning of his scientific career in the West Indies,
Dr. Watts has maintained a close contact between the chemical
laboratory and the Agricultural and Botanic Experiment Sta-
tions, and he has continued this policy of scientific cooperation
in all his subsequent administrative work. The effect of this
has been most beneficial, as results were secured which could
not have been accomplished had chemical, agricultural, botanical,
and industrial research proceeded along separate unassociated
lines.
The training of young students for the varied needs of indus-
trial life in the tropics is a subject to which the Imperial De-
partment of Agriculture has given much attention and a con-
siderable amount of Dr. Watt's time in late years has been de-
voted to questions of education. In addition to their usefulness
as centers of research, the experiment stations and laboratories
have been made to serve as training places where young students
may acquire practical first-hand knowledge of the subjects
taught in the elementary and secondary schools.
With the recent rapid growth which has taken place in de-
veloping the resources of the British West Indies a strong need
has been felt for a central higher institution of learning where
advanced students could obtain a complete theoretical and
practical training in the production of sugar, cacao, rubber,
and other agricultural commodities. The new Tropical Col-
lege, for which Sir Francis Watts has so long been working
and which is soon to be established in the island of Trinidad,
will remedy this need. Trinidad is an ideal location for the
new institution, for not only is it conveniently situated with
reference to the colonies in the West Indies and British Guiana,
but with its varied industries of sugar, cacao, rubber, limes, and
copra, as well as of asphalt and petroleum, it offers the student
almost unlimited natural facilities for study and research.
This college will be of much benefit to the Empire as a whole,
as well as to the colonies most immediately concerned, for up
to the present time the graduates of English universities who take
up scientific work in the tropics have lacked facilities for ac-
quainting themselves with the requirements of their new
duties.
The committee who have the matter in charge regard it as
desirable that an intimate relationship should exist between the
Tropical College and the Imperial Department of Agriculture,
and have recommended that the first president of the new in-
stitution should be the Imperial Commissioner of Agriculture.
The wide experience of Sir Francis Watts in the agricultural,
industrial, and educational life of the West Indies is sufficient
proof of the wisdom of this recommendation. While the ad-
ministrative duties of Sir Francis have obliged him to withdraw
from active work in -the laboratory, his original interest in
chemistry has continued unabated, and it is safe to predict
that under his leadership chemical research, as a means of
developing the industrial and agricultural resources of the trop-
ics, will find an important place in the curriculum of the new
college.
Sir Francis Watts by visits and by correspondence has always
kept in close touch with the work of his scientific confreres in
the United States, as well as in other parts of the world. He
has been a visitor at the Chemists' Club in New York, and those
who have met him there recall with pleasure his charming cordial
personality. His fellow members of the American Chemical
1 Society not only congratulate him for his enduring accomplish-
f ments but extend to him their best wishes for long years of
^.helpful activity to come.
RESEARCH PROBLEMS IN COLLOID CHEMISTRY
By Wilder D. Bancroft
rnell University, Ithaca, N.
Received November 5, 1920
The following list of problems was compiled at the request
of Prof. H. N. Holmes, Chairman of the Committee on the
Chemistry of Colloids of the Division of Chemistry and Chemical
Technology of the National Research Council. I have received
valuable assistance in preparing this list from Messrs. Holmes
and Weiser.
The arrangement is somewhat arbitrary because almost any
one of the problems could have been entered under at least two
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
heads, depending upon the particular aspect of the problem
that interested one; but a poor classification is distinctly better
than none at all. It is hoped that the publication of this list
will stimulate research in colloid chemistry. The committee
will be glad to receive suggestions as to additional problems.
In order to keep in touch with what is being done in this country
and in order to prevent unnecessary duplication of effort, the
committee will appreciate it if anybody who starts work on
any of these problems will send word to that effect to Prof.
H. N. Holmes, Oberlin, Ohio, who will furnish additional in-
formation, if desired, and who will also have copies of the list
for distribution.
ADSORPTION OF GAS OR VAPOR BY SOLID
(i) PRESSURE-CONCENTRATION ADSORPTION CURVES FOR HIGH
PRESSURES — It is believed that the adsorption isotherm for gases
has the same general form as the adsorption isotherm for solu-
tions and that at high pressures the adsorption varies very little
with increasing pressure. Dewar1 claims to have obtained an
isotherm of this type with hydrogen in charcoal at — 185°;
but he finds an adsorption of 156, 149, 145, and 138 cc. per gram
for pressures of 10, 15, 20, and 25 atmospheres, respectively,
and these adsorptions are not strikingly constant. At ordinary
temperatures and pressures the adsorption isotherm for hydrogen
in charcoal is nearly a straight line.2 Richardson3 gets approxi-
mately the theoretical curve for ammonia in charcoal at • — 64 °
and nearly a straight line at +1750. While there is no doubt
but that the nearly linear curves bend round at higher pressures,
this should be proved experimentally.
(2) ADSORPTION ISOTHERM FOR CO2 ABOVE AND BELOW THE
critical temperature — Mitscherlich4 calculated that, when
carbon dioxide at atmospheric pressure and 12° is adsorbed by
boxwood charcoal, the carbon dioxide occupies only one fifty-
sixth of its original volume. Since this is a lesser volume than
the same amount of carbon dioxide can occupy as a gas at this
temperature it is usually assumed that part has liquefied. This
assumption is the more probable because the heat of adsorption
of a gas or vapor is always somewhat larger than its heat of
liquefaction* It has been pointed out, however, by Mr. Johns-
ton that an adsorbed gas may be in such a state that it does not
liquefy even when compressed into a volume which it could not
occupy as gas in the free state. It is difficult to account for the
heat of adsorption on this view. The best way to test this
hypothesis would seem to be to determine adsorption isotherms
for carbon dioxide at temperatures above and below its critical
temperature, and at pressures up to those at which it would
liquefy in absence of charcoal. It is quite possible that these
experiments would throw some light on the form of the adsorp-
tion isotherm as discussed in No. 1. If Richardson's results
with carbon dioxide were plotted on a different scale, they might
answer the question.
(3) DATA TO SHOW THAT THE ORDER OF ADSORPTION OF GASES
AND VAPORS IS NOT NECESSARILY THAT OF THE BOILING POINTS —
It is often stated as a first approximation that a gas or vapor is
adsorbed more readily the higher its boiling point. Thus, helium
is adsorbed by charcoal much less than hydrogen, and hydrogen
again is adsorbed to a much less extent than nitrogen or oxygen.
Carbon dioxide is adsorbed less readily than ammonia, so these
substances follow the empirical rule. Argon, however, is ad-
sorbed less completely by charcoal than is nitrogen, while car-
bon monoxide is adsorbed to a greater extent at o° than either
argon or oxygen, though this is not according to the rule. Nitrous
oxide is adsorbed less strongly than ethylene, and nitric oxide
' Proc. Roy. Inst., 18 (1906), 437.
» Titoff, Z. physik. Chem., 74 (1910), 641.
» J. Am. Chem. Soc, 38 (1917), 1828.
< Sits. Akad. Wiss. Berlin, 1841, 376.
'Favre, Ann. chim. phys., [5] 1 (1874), 209; Lamb and Coolidge,
J. Am. Chem. Soc., 43 (1920), 1146.
more strongly than methane, which is not according to the boiling
points. Ethane, ethylene, and acetylene are adsorbed more
at +200 than is carbon dioxide, though the last is the most
readily condensable gas of the four. The difference between
carbon dioxide and hydrogen sulfide is in the right direction,
but seems out of all proportion to the difference in boiling points.
Hydrogen sulfide is adsorbed more than ammonia, although the
two boiling points are practically identical. Cyanogen is ad-
sorbed more than ammonia at 70° and less at 0°. In the case
of vapors there is no apparent relation between boiling point
and adsorption by charcoal. Going from higher to lower boiling
points, we have the order: water, benzene, ethyl alcohol, carbon
tetrachloride, methanol, chloroform, ether, and acetaldehyde.
The order from greater to lesser adsorption is: ethyl alcohol,
methanol, acetaldehyde, ether, benzene, water, chloroform and
carbon tetrachloride.1 There should be a systematic study of
the relations so that comparisons could be made at corresponding
temperatures and pressures. At temperatures below the critical
temperature, the limiting adsorption depends only on the pore
space and on the amount of contraction which the adsorbed
liquid undergoes.
(4) REPETITION OF HUNTER'S EXPERIMENTS ON THE ADSORP-
TION OF GASES BY DIFFERENT CHARCOALS AFTER TREATMENT
with steam AT 250° — Hunter2 found that charcoals made from
different woods behaved differently. The coconut charcoal had
the greatest adsorbing power of all. Of the others, charcoal
from logwood was the best with ammonia, charcoal from fustic
the best with carbon dioxide, and charcoal from ebony the
best with cyanogen. These results should be checked to make
sure that they are correct. The varying relative adsorption of
different gases by different charcoals is probably due at least
in part to the presence of different adsorbed impurities which
affect the different gases differently. The different charcoals
should be treated with steam at 250 ° to 300 ° in order to remove
as much as possible of the adsorbed impurities, and should then
be tested again.
(5) THE ADSORPTION OF AMMONIA BY AMMONIUM HYDRO-
sulFide — Magnusson3 found that the adsorption of ammonia
by ammonium hydrosulfide was sufficient to introduce a serious
error into the determination of the equilibrium relations for
ammonia and hydrogen sulfide. The problem should now be
reversed and a study made of the adsorption of ammonia by a
porous mass of ammonium hydrosulfide.
(6) study of vapor pressure curves of adsorbed water — ■
We get rather curious results if we apply Hatschek's view4 on
viscosity to Bingham's experiments5 on zero fluidity. If we
make the assumption that plastic flow is reached when the sur-
faces of adsorbed water are in contact, and if we make the further
assumption that we are dealing with spheres in open piling, the
voids will then be 48 per cent of the whole, and in the case of
graphite, for instance, the amount of water adsorbed by the
graphite must be 94.5 ■ — 48 = 47.5 volume per cent, or each
volume of graphite must adsorb about nine volumes of water.
If we assume close piling or different sizes of graphite powder,
the voids will be less and the amount of water to be adsorbed
will be greater. Since the volumes of two spheres are propor-
tional to the cubes of the radii, one volume of graphite will
hold seven volumes of water if the thickness of the water film
is equal to the radius of the graphite particles. If the thickness
of the water film is 1.2 times the radius, the graphite will hold
eleven volumes of water. This is the same type of calculation
'Hunter, Phil. Mag., [4] 26 (1863), 364; J. Chem. Soc, 18 (18651,
285; 20 (1867), 160; 21 (1868), 186; 23 (1870), 73; 24 (1871), 76; 26 (1872),
649; Dewar, Proc. Roy. Soc, 74 (1904), 124; Hempel and Vater, Z. Elek-
trochem., 18 (1912), 724.
' Phil. Mag., [41 26 (1863), 364.
3 J. Phys. Chem., 11 (1907), 21.
« Z. Kolloidchem., 11 (1912), 280.
> J. Frank. Inst., 181 (1916), 845.
Jan., 1921
TEE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
$S
that Hatschek made, and it shows that it is theoretically possible
on this assumption to account for zero fluidity in a graphite-
water mixture containing 5.5 per cent graphite. It has not been
shown, however, that the adsorbed films of water on the graphite
particles are of the desired thickness, nor has it been shown
that 47 per cent of the water in the mixture is in a different
state from the rest of the water. It might be possible to do
this last by measuring the vapor pressure curve for graphite-
water mixtures and determining the point at which the vapor
pressure became that of pure water.
(7) APPARENT VOLUME OF POWDERS IN A VACUUM — As little
as 5 per cent of the apparent volume of a mass of carbon black
may be due to the solid,1 and a liter of carbon black may contain
2.5 liters of air.2 If the adsorbed air were all pumped out, the
apparent volume of the carbon black would undoubtedly be
very much less; but nobody has actually proved it. An experi-
ment to prove this would be interesting because it would furnish
a new proof of the existence of the film of adsorbed air. It
is also important to know the true voids in a mass of carbon
black or other substance, because this value plays an important
part in the theory of viscous and plastic flow as developed by
Bingham.3 Still more striking results could probably be ob-
tained by working in an atmosphere of carbon dioxide or of
ammonia, especially if powdered charcoal were substituted for
carbon black.
When indigo is reduced to a very fine powder by means of a
disintegrator,4 the single particles appear to be separated one
from another by an envelope of air, so that the dry powder occu-
pies only 20 per cent of the apparent volume. Cushman and
Coggeshall' found that cement rock powder which would pass
through a 200-mesh sieve surged like a liquid because of the film
of adsorbed air. When poured into a vessel the fine powder
filled only 46 per cent of the space, while a coarser powder filled
more. Finely ground phosphate rock also flows like a liquid. In
all these cases pumping out the adsorbed air would undoubtedly
make the powders pack more closely, but this has not yet been
proved experimentally.
(8) EFFECT OF COMPRESSING POWDERS IN PRESENCE OF AD-
SORBED gas — Platinum black takes up a great deal more hydrogen
than does platinum foil. If the hydrogen were dissolved in the
platinum the equilibrium concentrations would be the same in
both cases. While it is probable that some hydrogen is dissolved
in the platinum, it is difficult to tell how much because of the
slowness in reaching equilibrium. IPwe start with a platinum
black saturated with hydrogen, and burnish the platinum black
without removing it from the hydrogen, any hydrogen which is
set free will be adsorbed hydrogen, and a measurement of the
amount will give some clue as to the relative amounts of dis-
solved and adsorbed hydrogen in the platinum. Similar ex-
periments should also be made with palladium and hydrogen.
If powdered alumina or other material is compressed to a
solid mass in presence of an adsorbed gas, much of the adsorbed
gas will be set free and none of the dissolved gas in case any is
present.
(9) ADSORPTION ISOTHERMS FOR MIXTURES OF GASES — In
many cases the adsorption of one gas by a solid decreases the
amount of a second gas which can be adsorbed ; but there are no
satisfactory quantitative measurements to show this.* Ad-
sorption isotherms should be determined, showing the relative
amounts of two gases in the vapor phase and in the charcoal
phase when in equilibrium at constant pressure.
(10) BEHAVIOR OF MIXTURES OF CARBON BISULFIDE AND
illuminating gas with coconut charcoal — According to
1 Cabot, 8th Inlernat. Congr. Applied Chemistry, 12 (1912), 18.
* Sabin, "Technology of Paint and Varnish," 1917, p. 201.
' Am. Chem. J., 46 (1911), 278; J. Frank. Inst., 181 (1916), 845.
' J. Soc. Dyers Colourists, 17 (1901), 294.
'J. Frank. Inst., 174 (1912), 672.
J Hempel and Vatcr, Z. Eleklrochem., 18 (1912), 724
Matwin1 charcoal will take carbon bisulfide and carbonyl sul-
fide out of illuminating gas, one kilogram of charcoal cutting the
sulfur content of 10 cubic meters of gas to 2.92 g. Porous
charcoals are the best, such as pine and linden. Bone-black
takes up almost no carbon bisulfide, and coconut charcoal is
said to be even less effective. This seems very remarkable be-
cause coconut charcoal adsorbs carbon bisulfide strongly. If
the statement is correct, the illuminating gas must cut down the
adsorption of carbon bisulfide very much. If carbon bisulfide
and illuminating gas were adsorbed in the same ratio in which
they occur in the mixture, an analysis of the gas coming through
would show an apparent purification2 even though the total
adsorption were very large.
(il). DOES THE EFFECT OF A TEMPERATURE GRADIENT ON THE
MOVEMENT OF SMOKE PARTICLES DEPEND ON THE NATURE OF
THE SMOKE PARTICLES AND OF THE SURROUNDING GAS? —
Aitken3 has shown that a suspended smoke particle moves along
a temperature gradient from the hotter to the colder portion.
If this is due to the presence of an adsorbed gas film around
the smoke particles, the phenomenon must vary quantitatively
with the nature and physical state of the smoke particle and
with the nature of the gas. As yet there are no experiments to
prove this.
(12) DO ELECTRICAL WAVES OR STRESSES HAVE A MEASURABLE
EFFECT ON the adsorption of gases? — Schuster4 pointed out
that some of the most puzzling facts of the disruptive discharge
admit of explanation if we assume the existence in contact with
the electrode of a surface layer of condensed gas having a large
inductive capacity. If the layer of adsorbed gas offers an in-
creased resistance to the passage of an electrical discharge, it
follows from the theorem of LeChatelier that an electrical
stress will tend to remove the film of adsorbed gas. This enables
us to account for many apparently unrelated facts in connec-
tion with over-voltage, with colliding drops, and with the elec-
trolytic detector, the crystal detector, and the coherer as used
in wireless telegraphy.6 While this point of view has proved
useful, its accuracy has never been demonstrated experimentally.
It is very desirable that we should have experimental proof that
electrical waves or stresses do decrease the adsorption of gases.
(13) decomposition of sodium amalgam — Fernekes6 found
that alcohol and many other organic substances increased the
rate of reaction between sodium amalgam and water. He
accounts for the phenomenon by assuming the intermediate
formation of hypothetical compounds between solvent and
solute which are extremely unstable towards sodium amalgam
and, therefore, react very rapidly with it. While this explana-
tion may be right, it has not proved helpful and is, therefore,
useless, at any rate for the present. It seems probable that
certain organic substances lower the over-voltage at mercury,
and consequently make, the sodium amalgam unstable. This
hypothesis is susceptible of proof by direct experiment. While
there are no measurements as yet made under conditions strictly
comparable to those in Fernekes' experiments, Carrara7 has
shown that the over-Voltages are quite different in methanol
and in ethyl alcohol from what they are in water. I have
often wondered whether the reason that nobody has ever pre-
pared, electrolytically, a sodium alloy using a cathode of fused
Wood's alloy, might be because the over-voltage is not sufficient
in this case.
(14) fixation OF oxygen by carbon — Rhead and Wheeler8
discuss the adsorption of oxygen by carbon as follows:
■ J. Gasbel., 62 (1909), 602.
* Cf. Leighton, J. Phys. Chem., 20 (1916), 32.
' Trans. Roy.. Soc. Edinburgh, 32 (1884), 239; Bancroft, J. Phys.
Chem., 24 (1920), 421.
« Phil. Mag., [51 29 (1880), 197.
« Bancroft, J. Phys. Che,m., 20 (1916), 18, 402, 503.
'Ibid., 7 (1903), 611.
' Z. physik. Chem., 69 (1909), 75.
•/. Chem. Soc, 103 (1913), 462.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
The experiments show that carbon, at all temperatures up
to 900 ° and probably above that temperature, has the power
of pertinaciously retaining oxygen. This oxygen cannot be
removed by exhaustion alone, but only by increasing the tem-
perature of the carbon during exhaustion. When quickly re-
leased in this manner it appears, not as oxygen, but as carbon
dioxide and carbon monoxide. The proportions in which it
appears in these two oxides when completely removed depend
on the temperature at which the carbon has been heated during
oxygen fixation. No physical explanation alone can account
for this fixation of oxygen; but, in all probability, it is the out-
come of a physicochemical attraction between oxygen and car-
bon. Physical, inasmuch as it seems hardly possible to assign
any definite molecular formula to the complex formed, which,
indeed, shows progressive variation in composition; chemical,
in that no isolation of the complex can be effected by physical
means. Decomposition of the complex by heat pioduces carbon
dioxide and carbon monoxide. At a given temperature of de-
composition these oxides make their appearance in a given ratio.
Further, when a rapid stream of air at a given temperature is
passed over carbon (which has previously been "saturated"
with oxygen at that temperature) carbon dioxide and carbon
monoxide appear in the products of combustion in nearly the
same ratio as they do in the products of decomposition of the
complex at that temperature. Our hypothesis is that the first
product of combustion of carbon is a loosely formed physico-
chemical complex, which can be regarded as an unstable com-
pound of carbon and oxygen of an at present unknown formula,
CjOy. It is probable that no definite formula can be assigned
to this complex.
It is perfectly possible that the mysterious oxide is a definite
compound which is adsorbed by the charcoal and which, there-
fore, has a decomposition pressure1 which varies with varying
temperature. On this hypothesis the pure compound, possibly
CisO«, or a decomposition product,8 perhaps a compound3
C»0, would behave in one way when heated by itself and quite
differently when adsorbed by charcoal. Decomposition pres-
sures and compositions should be determined for mellitic acid,
the oxide C12O1, and any other compound, oxalic acid for instance,
which might conceivably break down to form a compound having
the properties described by Rhead and Wheeler. First-class
charcoal should then be impregnated with these substances and
the experiments repeated. It is not necessary to assume that
the compound breaks down in different ways at different tem-
peratures. There is always an excess of carbon present, and,
on slow heating, one would probably always come very close
to the equilibrium ratio for carbon dioxide, carbon monoxide,
and carbon for the temperature in question. If a current of an
inert gas were passed rapidly through the system so as to sweep
out the decomposition products as fast as formed, it ought to
be possible to approximate to the decomposition products which
the compound would give if heated by itself.
(15) oxidation temperature for carbon — The experiments
of Manville* on the oxidation of carbon were undoubtedly vitiated
by the presence of hydrocarbons. These experiments should be
repeated with charcoal which has been freed from hydrocarbons
by treatment with steam.
(16) synthesis of mellitic acid — The experiments of Meyer6
seem to show that pure carbon cannot be oxidized to mellitic
acid and that the mellitic acid obtained by the oxidation of
ordinary wood charcoal is due to the oxidation of some hydro-
carbon. To make the proof conclusive, it ought to be shown
what hydrocarbons oxidize to mellitic acid under the conditions
of the experiment. With our modern technique, this should not
be difficult.
(17) determination of heats of adsorption — We have
1 Bancroft, /. Phys. Client., 24 (1920), 220.
2 Diels and Wolf, Ber., 39 (1906), 689; Diels and Meyerheim, Ibid.. 40
(1907), 355; Meyer and Steiner, Ibid., 46 (1913), 813; Armstrong and Cole-
gate, J. Soc Chem. Ind., 32 (1913), 396.
» Lowry and Hulett, J. Am. Chem. Soc, 42 (1920), 1408.
* J. Mm. phys., 6 (1907), 297; Duhem, Van Bemmelen Cedenkboek,
1910, 1; Lowry and Hulett, J. Am. Chem. Soc, 42 (1920), 1408.
» Monatsh., 35 (1914), 163.
very few measurements on the heats of adsorption of gases,1
and some of these are not very accurate. The subject is an
important one2 and measurements should be made with great
accuracy. The heats of adsorption of hydriodic acid and of
hydrobromic acid by charcoal are several times the latent
heat of vaporization, and we do not know at all why the molecu-
lar heat of adsorption of hydrogen should be 18,000 calories
with palladium and about 46,000 calories with platinum.
contact catalysis
(18) effect of co adsorption, etc., on adsorption op
hydrogen, ethylene, ETC. — We know that carbon monoxide
cuts down the catalytic action of platinum8 on hydrogen and
ethylene, and we believe that this is because it cuts down the
adsorption of these gases; but there are no satisfactory quanti-
tative measurements on the adsorption by platinum of mixtures
of CO with hydrogen or ethylene. Maxted* has made some
measurements on hydrogen sulfide and hydrogen with palladium.
(19) ADSORPTION BY, COLLOIDAL PLATINUM OF SUBSTANCES
which poison hydrogen peroxide — While we are quite certain
that the poisoning of the platinum catalysis of hydrogen peroxide4
is due to the adsorption of the so-called poisons, there are not
even qualitative experiments to prove this. Platinum black
should be shaken with solutions of the different poisons and ad-
sorption isotherms determined.
(20) behavior of potassium cyanide solution with col-
loidal PLATINUM, PLATINUM BLACK, AND MASSIVE PLATINUM —
Bredig6 points out that when colloidal platinum is allowed to
stand in contact with hydrogen peroxide and concentrated
potassium cyanide, the platinum flocculates and precipitates.
The agglomerated platinum causes the hydrogen peroxide to
decompose, thus showing that the cyanide does not poison pre-
cipitated platinum black. There seem to be only two possible
explanations. One is that the adsorption of potassium cyanide
by platinum falls off very much more rapidly with increasing
size of the platinum particles than the adsorption of hydrogen
peroxide by platinum. The other explanation is that, through
oxidation or otherwise, there is formed what might be called
an anti-body, which cuts down the adsorption of the cyanide.
Neither hypothesis is very satisfactory and there is no experi-
mental evidence for either. This point should be cleared up.
Kastle and Loevenhart7 point out that prussic acid accelerates
the decomposition of the hydrogen peroxide by iron and copper.
There is no theory in regard to this.
(2 I ) APPARENT EQUILIBRIUM BETWEEN PHOSGENE AND AQUEOUS
hydrochloric acid — Phosgene reacts with water to give
carbon dioxide and hydrochloric acid:
COCh + H20 = CO2 4- 2HCI
So far as we know, this reaction is not reversible, and it ac-
tually runs to an end in presence of an excess of water. In
presence of concentrated hydrochloric acid the rate of hydrolysis
is practically negligible. The only way that I can see to ac-
count for this is by assuming that water and phosgene do not
react by themselves and that the reaction takes place solely
in contact with the walls of the containing vessel. When these
are coated with a film of hydrochloric acid of sufficient concen-
tration, no phosgene is adsorbed to speak of, and no reaction
takes place. The hydrolysis should be studied with different
concentrations of acid and with a varying ratio of wall surface
to mass of solution.
1 Favre, Ann. chim. phys., [5] 1 (1874), 209; Masson, Proc. Roy. Soc,
74 (1904), 209; Dewar, Proc. Roy. Inst., 18 (1905), 183.
* Lamb and Coolidge, /. Am. Chem. Soc, 42 (1920), 1146.
« Lunge and Harbeck, Z. anorg. Chem., 16 (1898), 50.
« J. Chem. Soc, 118 (1919), 1020.
» Bredig and von Berneck, Z. physik. Chem., 31 (1899), 258; Bredig
and Ikeda, Ibid., 37 (1901), 1.
« Z physik. Chem., 31 (1899), 332.
' Am. Chem. J., 29 (1903), 397.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
87
(22) EFFECT OF OSCILLATING TEMPERATURES ON THE AP-
PARENT EQUILIBRIUM OF ETHYL BUTYRATE WITH LIPASE — Tri-
chloromethyl chloroformate, CICO2CCI3, or superpalite as it
has been called, decomposes to carbon tetrachloride and carbon
dioxide in presence of alumina,
C1C02CC1S = C02 + CCU,
and to phosgene in presence of ferric oxide,
C1C02CC13 = 2COCI2.
The reverse reaction has never been made to take place to
any measurable extent. Some superpalite and ferric oxide were
placed in a glass tube connected with a closed manometer.
There was rapid decomposition at first, as shown by the in-
crease in pressure; but, before long, the reaction came apparently
to an end. On raising the temperature the reaction went a
little farther and did not reverse when the temperature was
brought back to its original value. This experiment was not
checked sufficiently to make me willing to guarantee the results ;
but it looks as though the ferric oxide was poisoned and that
when the temperature changed, more superpalite came in con-
tact with the catalytic agent and was decomposed. If this is
the true explanation, it suggests one interesting line of experi-
mentation. When ethyl butyrate is treated with a small amount
of enzyme, the decomposition proceeds only a little way.1 It
seems probable that with an oscillating temperature it might
be possible to carry the reaction much farther with the same
amount of enzyme.
(23) ACTION OF PLATINUM BLACK ON ACETIC ACID — Reiset
and Millon2 state that acetic acid can be boiled with pumice
without decomposition; but that it is decomposed completely
if distilled from platinum black. They do not state what the de-
composition products are. At first there might be enough oxygen
in the platinum black to cause an oxidation of the acetic acid;
but that would soon come to an end. We are not absolutely
certain that platinum black does decompose acetic acid catalyt-
ically at the boiling point of the latter. If that does happen,
we can only guess at the reaction products.
(24) CATALYSIS OF ETHYL ACETATE IN PRESENCE OF HYDROGEN
— If a mixture of ethyl acetate vapor and hydrogen is passed
over pulverulent nickel, it is probable that some or all of the
initial products will be reduced before they have time to react
in the normal way. A study of the reaction products should,
therefore, throw light on the probable mechanism of the reaction
which occurs in the absence of hydrogen. If methane and ethyl
formate are the products, that would indicate that the original
break had been into -CH3 and -CO2C2H5. If acetic acid and
ethane are found, they would probably be reduction products
of CH3CO2- and -CH2CH3. If the reaction products are me-
thane, ethane, and either carbon dioxide or some of its reduc-
tion products, it would seem certain that ethyl acetate splits
simultaneously into -CH3, -CH2CH3, and C02.
(25) catalysis OF ETHER by nickel — If ether is passed over
pulverulent nickel, one stage in the reaction will probably
be to CH3CH2O- and -CH2CH3 or to C2H6OC2H4- and -H. In
the first case the final products will be ethylene and water just
as with alumina. In the second case they are likely to be
acetaldehyde, ethylene, and hydrogen, though the ethylene and
hydrogen may combine more or less completely to form ethane.
A study of this reaction should, therefore, throw light on the
catalytic decomposition of alcohol by nickel.
(26) CATALYSIS OF METHYL FORMATE BY ALUMINA AND FERRIC
oxide — We have data for the catalytic decomposition of tri-
chloromethyl chloroformate by alumina and by ferric oxide. As
soon as we get the corresponding data for methyl formate, we
shall be in a position to tell whether the substitution of hydrogen
by chlorine changes the type of the reaction.
' Kastle and Loevenhart, Am. Chem. J., 21 (1900), 491.
2 Compt. rend., 16 (1843), 1190.
(27) catalytic action of ferrous oxide — Since alumina is
very transparent and ferrous oxide very opaque to infra-red
radiations, ferrous oxide should be much superior to alumina as
a catalytic agent, according to the radiation theory of W. C.
McLewis, in all cases where the formation of metallic iron or
of another oxide did not interfere with its activity.
(28) gum Arabic as catalytic agent— According to Tyndall,1
gum arabic is practically opaque to infra-red rays. If this is so,
it must emit infra-red rays and should, according to the radia-
tion theory, be a powerful catalytic agent for methyl acetate
solutions. This would seem to be a crucial experiment.
(29) ARSENIC POISONING OF THE GRILLO-SCHROEDER CONTACT
mass — The Grillo-Schroeder catalyst for the contact sulfuric
acid process consists of platinum black precipitated in a certain
way on magnesium sulfate. This contact mass is poisoned by
arsenic just as is the platinized asbestos. It has been stated,
however, that the Grillo-Schroeder catalyst can be regenerated
by boiling with hydrochloric acid. It was supposed that the
arsenic was removed as trichloride ; but analysis showed that the
regenerated contact mass contained a great deal of arsenic. The
amount was said to be 3 per cent, but I do not know whether
this was 3 per cent of the amount of platinum or of the contact
mass. This arsenic must either have agglomerated, so that it
no longer coated the platinum, or it must have reacted with the
magnesium sulfate. It might be very difficult to tell from a
microscopic examination what had happened, so that it probably
would be better to study first the behavior of arsenic with porous
magnesium sulfate in the absence of platinum.
(30) SPONTANEOUS COMBUSTION OF OILED RAGS — It is known
that oiled rags will take fire spontaneously, and there is some litera-
ture on the subject.2 In view of the number of fires which seem
to be due to this cause, somebody ought to develop a really
first-class lecture or laboratory experiment tc illustrate this, and
the experiment should be included in every introductory course
in chemistry.
(31) IGNITION TEMPERATURE OF GAS MIXTURES — When gas
mixtures are exploded by an incandescent wire or by a spark,3
it seems probable that the nature of the wire or of the electrode
has a catalytic effect, at any rate at the outset. If this is the
case, it should be possible to poison the wire to some extent.
Presence of carbon monoxide might perhaps change the apparent
ignition temperature for oxyhydrogen gas. Something of this
sort might account for the change in temperature when the
mixture is diluted with one of the constituents and for the effect
of sparks which do not cause explosion.
(32) DECOMPOSITION OF VERMILION BY COPPER — De la Rue4
states that electroplated copper blocks cause vermilion to blacken,
while cast copper does not. If this is true, the difference must
be due to the greater porosity of the electroplated copper. The
matter should be tested, so that we may know the facts.
ADSORPTION OF VAPOR BY LIQUID
(33) COALESCENCE OF COLLIDING DROPS OF DIFFERENT
Liquids — -Lord Rayleigh6 has shown that colliding drops or
jets of water do not necessarily unite. This is because of a
film of adsorbed air which prevents the drops from coming ac-
tually in contact. This phenomenon must be general, and must
be most marked the greater the adsorption of gas by the liquid
drops. Experiments should, therefore, be made with drops of
nonaqueous liquids and in different atmospheres. It has also
1 "Fragments of Science," "Radiant Heat and Its Relations."
2 Galletly, Chem. Zentr., 1873, 543; Coleman, J. Chem. Soc, 31 (1878),
259; Kissling, Z. angew. Chem., 1896, 44; Lippert, Ibid., 1897, 434.
* Roszkowski, Z. physik. Chem., 7 (1896), 485; Coward, Cooper and
Warburton, J. Chem. Soc., 101 (1912), 2278; Parker, Ibid., 106 (1914), 1002;
Sartry, Ibid., 109 (1916), 523; McDavid, Ibid., Ill (1917), 1003; White
and Price, Ibid., 116 (1919), 1462; Thornton, Proc. Roy. Soc., 90A (1914),
272; 91A (1914), 17; 92A (1915), 9, 381; Phil. Mag., [6] 38 (1919), 613.
* Mem. Chem. Soc, 2 (1845), 305.
« Proc. Roy. Soc, 28, 406; 29 (1879), 71; 31 (1882), 130: Bancroft,
J. Phys. Chem., 20 (1916), 1.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
been shown by Lord Rayleigh1 and others* that an applied po-
tential difference of about two volts will cause colliding drops
to coalesce; but this value has not been determined accurately,
and we do not know how it woidd vary, if at all, with solutions
instead of so-called pure water. Both these matters should be
studied.
(34) STUDY OF ORNDORFF AND CARRELL'S EXPERIMENTS ON AIR-
BUBBLING — In some experiments with the air-bubbling method
of determining molecular weights, Orndorff and Carrell3 found
that with urethane solutions approximately theoretical values
were obtained even when the rate of bubbling was varied a
great deal. With urea solutions there is a distinct tendency for
the apparent molecular weight to go up as the rate of bubbling
is increased. With phenol the apparent molecular weights were
low at all rates of bubbling and did not vary much with the rate
of bubbling. The experiments of Campbell4 make it probable
that some of the errors in the air-bubbling method are due to the
presence of an adsorbed gas film on the surface of the liquid.
The experiments of Orndorff and Carrell should be repeated,
amplified, and studied with special reference to the work of
Campbell. These same solutions might well be tried in No. 33.
(35) EFFECT OF POWDERS IN MAKING DROPS COALESCE — Lord
Rayleigh5 found that dry powders had a marked effect in causing
colliding drops or jets of water to coalesce, whereas most of
the powders were ineffective when wetted. No explanation
was^ given for the phenomenon and yet one should be found.
It is possible that the electrification of the powders may be a
factor. Hardy6 has noticed that powders floating on a liquid
sometimes move in the opposite direction from the same
powders when submerged.
ADSORPTION OF LIQUID BY SOLID
HNORMAL DENSITY OF POWDERS IN LIQUIDS — Rose7
claims that platinum in the state of foil has a specific gravity
of 21 to 22, while a value of about 26 was obtained for platinum
sponge precipitated from the chloride by sodium carbonate and
sugar. This can be accounted for if we assume that the powder9
is not weighed alone in water, but in conjunction with a film
of condensed water. Similar, though less extreme differences
were obtained with gold, silver, and barium sulfate. These
experiments should be repeated and extended.
(37) SYSTEMATIC STUDY OF RELATIVE WETTING, WITH SPECIAL
REFERENCE TO FLOTATION AND TO ZERO FLUIDITY — No System-
atic study of the selective adsorption of liquids by solids
seems to have been made. There are a few scattered data.
We know that kerosene will displace water in contact with metals,
and that water will displace kerosene in contact with quartz.9
while alcohol will displace oil in contact with metal,10 and linseed
oil11 will displace water in contact with white lead. When
making lithographic inks, oil is added to the wet paste and the
water is ground out. There are only a few quantitative measure-
ments15 on the selective adsorption of a liquid by a solid. A
careful systematic study of the phenomenon should be made.
It is the determining factor in ore flotation. If we get zero
fluidity15 when the voids in a powder are just filled with liquid,
' Proe. Roy. Soc, 29 (18791. 7 1 .
= Newall, Phil. Mag., [5] 20 (1885), 31; Burton and Wiegand, Ibid.,
23 (1912). 14S.
' J. Phys. Chcm., 1 (1897), 753.
' Trans. Faraday Soc, 10 (1915), 197.
' Proc. Roy. Soc, 31 (1882), 130; Bancroft, J. Phys. Chem., 20 (1916), 14.
Joe, 86A (1912). 609.
I Pogg Auk., 73 (1848), 1; J. Chem. Snc, 1 (1849). 182.
s See, however, Johnston and Adams, /. Am. Chan. Soc, 31 (1912),
563.
' Hofmann, Z. physik. Chcm., 83 (1913), 385.
"> Pockels, Wild. Ann., 67 (1899), 669.
II Cruickshank Smith, "The Manufacture of Paint," 1916, p. 92.
'-Graham. J. Chem. Soc, 20 (1867), 275; Mathers, Trans. Am. Elec-
trochem. Soc, 31 (1917), 271.
" Bingham, ,4m. Chem J., 46 (1911), 278; J. Frank. Inst., 181 (1916),
845.
the extra liquid is present as an adsorbed film and the determina-
tion of the amount is very important.
(38) BEHAVIOR OF GUM ARABIC WITH ALCOHOL AND WATER — ■
It is not very easy to peptize gum arabic by grinding with water
because the water does not displace the air readily from the gum.
If the gum is ground for a moment with alcohol, water then
wets it readily. This is surprising because water peptizes the
gum and alcohol does not; one would consequently have ex-
pected the water to be adsorbed more strongly than the alcohol.
By shaking the gum arabic with aqueous alcohol, it should be
an easy matter to tell whether the alcohol or the water is ad-
sorbed the more strongly. It is possible that there may be a
film of grease on the gum which is removed by the alcohol.
It is possible that alcohol displaces the air more rapidly because
it adsorbs the air more strongly than does water. If that is
the case, alcohol should show a special behavior as colliding
drops in No. 33. Experiments should be made with acetone,
acetic acid, glycerol, etc., so as to see to what extent the phe-
nomenon is general or to what extent it is peculiar to
alcohol.
We are always working up to the problem of why concen-
trated sulfuric acid wets sulfur trioxide more readily than water
does.
(39) BEHAVIOR OF MERCURY IN GLASS CAPILLARY AS AIR IS
removed — Mercury does not wet glass because air is adsorbed
more strongly than mercury by glass. According to this point
of view, mercury should wet glass if the air is removed com-
pletely. There are experiments by Hulett and others to show
that this is true; but the problem has never been handled in
a clear-cut manner. One would like to see mercury made to
rise in an evacuated glass capillary.
(40) carrying OF MERCURY on iron gauze — Lord Rayleigh1
pressed a piece of iron gauze down on the flat bottom of a glass
vessel holding a shallow layer of mercury, and found that the
gauze remained on the bottom of the vessel and did not rise
through the mercury. The reason for this is that the mercury
does not wet the iron. A corollary from this, which has not
been tested experimentally, is that one should be able to carry
mercury in an iron sieve just as one can carry water in an oiled
sieve.2 Since sodium amalgam wets iron,3 a dilute sodium amal-
gam should run through an iron sieve which would stop pure
mercury. Also Rayleigh's experiment should not succeed if
a sodium amalgam were substituted for mercury. All these
predictions should be confirmed or disproved experimentally.
(41) pressures due to selective wetting— When water
displaces air at the surface of a solid, one wonders how much
pressure might be developed. Jamin4 has made some prelim-
inary experiments along this line. A hole was bored in a piece
of dried chalk. Into this hole was dipped one end of a manometer,
and the hole was then closed. When the chalk was placed in
water, the air was displaced from the pores and a pressure of
3 to 4 atmospheres was obtained. This is not the maximum
pressure because the amount of dead space in the manometer,
was large. A better method would be to determine the pressure
necessary for the air to force the water out of the pores of the
chalk. It would also be interesting to substitute alcohol and
other liquids for water. By filling a porous block of silica with
kerosene and placing it in water, or by filling a porous block of
lead or zinc sulfide with water and putting it in oil, one could
measure pressures which might be of distinct interest in their
bearing on flotation and on oil deposits near the sea.
(42) constant-temperature baths — Mcintosh and Edson5
have frozen aqueous salt solutions in a mixture of ether and
1 Scientific Papers, i (1903), 430.
1 Chwolson, "Traite de Physique," 1, III (1907). 613.
> J. Chem. Soc, 26 (1873), 418.
< Chwolson, "Traite de Phj-sique." 1, III (1907), 622
' J Am. Chem Soc. 38 (1916). 613.
Jail.-, ^21
THE ThMRNAL of industrial and ENGINEERING CHEMISTRY
soM carbon dioxide. Trie solid mass is said to melt at a constant
temperature, that of the initial freezing point of the solution.
At present there is no theoretical explanation for this.
(43) THEORY OF adhEsives — The whole theory of adhesives
depends in part on the fact that the cementing material adheres
strongly to the two surfaces and hardens there. It is therefore
possible that one agglutinant may be useful for a number of
different materials, such as wood, glass, metal, ivory, etc.,
while others give good results only with special materials. Since
the books give different recipes for cements for glass, cements
for metals, cements for metals and glass, etc., the differences
in adsorption are real ones, though no one has ever made a
careful study of agglutinants from this point of view. Some-
body should study the different adhesives from this point of view.
(44) vegetable glues — There is practically no literature
on the vegetable glues outside of a few patents. We need
published research on the whole subject with special reference
to peptization, viscosity, and adsorption.
(45) waterproof GLUES — A waterproof glue of indefinite
life is needed. Our large timber is disappearing fast and, before
long, we shall be compelled to build up large pieces by gluing
together what we can get from small stuff. At present the
best waterproof glues weaken In time, no doubt because of the
action of water on the protein material. A glue should be made
that will not take up moisture after it has once dried.
(To be continued)
5CILNTIFIC 50CILTIL5
J
CROP PROTECTION INSTITUTE DISCUSSES WAR
ON BOLL-WEEVIL
A meeting of the Crop Protection Institute, recently organized
under the National Research Council and made up of growers,
scientists, and business men, was held at Rumford Hall, New
York City, on Monday, December 6, 1920.
The principal topic for discussion was the control of the
boll-weevil by the application of calcium arsenate. Cotton
growers have suffered great losses in recent years due to the rav-
ages of the boll-weevil, and although the Department of Agricul-
ture has worked out careful methods for combating this pest
by the use of calcium arsenate, the results have not always been
satisfactory owing to faulty technique in the application of this
chemical.
The attendance was made up of representatives of insecticide
manufacturers and of manufacturers of spraying machinery,
as well as the regular membership of the Institute.
Prof. B. C. Coad of the U. S. Agricultural Experiment Station
at Tallulah, La., who has done a great deal of work on the
control of the boll-weevil, presented a two-reel moving picture
entitled "Goodbye, Boll-Weevil" which demonstrated the com-
plete control that can be won over the insect by the proper use
of calcium arsenate with the right kind of machinery.
Professor Coad stated very plainly that there had been con-
siderable failure in the application of calcium arsenate in the
hands of persons who had been improperly informed on the
method of using it. He summed up the causes of failure as being
due to laxness in carrying out definitive instructions, bad chem-
icals, and misinformation passed on to the farmer by ignorant
salesmen. He also commented on the fact that many of the
dusting machines sold to users were inefficient.
In 1920, 10,000,000 lbs. of calcium arsenate had been sold
to the South, said Dr. Coad, but probably 5,000,000 lbs. re-
mained unused, owing to lack of results in many cases.
At one of the meetings of the scientists connected with the
Institute the problems involved in the production and use of
calcium arsenate were discussed at some length. The general
feeling was that a standard for total arsenic in commercial
calcium arsenate be prescribed and adhered to. The standard
which seemed most desirable was 40 to 42 per cent total arsenic.
In the discussion it was brought out that from five to seven times
the present annual consumption of arsenic in the United
States would be required for the control of the boll-weevil
alone.
The fact that about 115 scientific men and 23 commercial
concerns have already joined the Crop Protection Institute
and that the first real business meeting was so well attended
augurs well for its future. It was disappointing, however,
to the organizers to be informed by Dr. L. O. Howard in his ad-
dress that the scientists of the Federal Government did not
see their way clear to become members of the Institute evert
though they sympathized with its purposes. Although the
constitution provides for the control of the Institute by the
scientist members only, the government men feel that it would not
be proper to become actively identified with an organization,
the funds of which come largely from commercial sources.
AMERICAN INSTITUTE OF CHEMICAL ENGINEERS
The Thirteenth Annual Meeting of the American Institute
of Chemical Engineers was held in New Orleans, December
6 to 9, 1920. The meeting was held in New Orleans in order
to give opportunity to make a study of the characteristic in-
dustries of this section of the South. The program provided
for a stay of two and one-half days in New Orleans, a two-day
trip through the sulfur, salt, rice, and sugar region of the state
of Louisiana, and stops on the return trip at Chattanooga,
Tenn., Roanoke, Va., and Luray, Va. Arrangements had been
made at ail points visited for inspection of the local industries.
The program of papers contained several which were descriptive
of the local chemical industries.
Dr. R. F. Bacon presented a paper on "Recent Advances in
the American Sulfur Industry" in which he discussed the diffi-
culty encountered in burning Louisiana sulfur on account of
the presence of small amounts of petroleum.
Lezin A. Becnel presented a paper on "Operating Variations
in Sugar Production as Indicated by Some Plantation Data,"
in which the author gave the results of a study of the produc-
tion of sugar and sirup during a period of some 40 yrs.,
and contended that the greatest profits would be made by pro-
ducing either sugar or sirup, or both, according to the market
for each product. The paper was discussed by Professors
Chas. S. Williamson, Jr., of Tulane University, and Chas. E.
Coates, dean of the Audubon Sugar School.
A very interesting talk on the "Resources of the State of
Louisiana" was given by Mr. N. L. Alexander, chief of the State
Conservation Commission. Motion pictures of the extensive
state game preserves were shown. Mr. Alexander also described
very successful experiments in reforestation. It has been
demonstrated that timber suitable for wood pulp can be grown
in Louisiana in 15 yrs.
George G. Earle, chief engineer and superintendent of the
Sewerage and Water Board, described the sewage, water puri-
fication, and drainage systems of New Orleans. Particular
interest was shown in the low lift pumps used to raise the
storm waters and sewage of New Orleans to the level of the
water courses used for drainage.
The other papers presented were of a general chemical engi-
neering character. Most of them were fully illustrated by
lantern slides and were very fully discussed. They included:
90
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
E. R. Weidlein. The Conservation of Heat Losses as Applied to
Power and Heating Systems. (Lantern slides)
James R. Withrow and F. C. Vilbrandt. The Sulfuric Acid Fume
Problem.
A. G. Peterkin. Costs — A Short Study of Factory Economics.
(Lantern slides)
Maximilian Toch. Lubrication of Concrete. (Lantern slides)
E. Bartow. The Treatment of Sewage by Aeration in the Presence
of Activated Sludge. (Lantern slides)
James R. Withrow. The Federated American Engineering Societies
and the Institute.
C. B. Morey. The Salvaging of Sag Paste.
W. L. Badger. Studies in Evaporator Design. IV — Some Data
from the Horizontal Tube Evaporator.
During the stay in New Orleans a visit was made to the plant
of the U. S. Industrial Alcohol Company where molasses is
diluted and fermented, and 95 per cent alcohol distilled out.
On the same afternoon the water purification plant of the city
of New Orleans was visited.
The plants of Pennick and Ford, as well as that of the
Southern Cotton Oil Co., were not visited as originally
planned on account of a very severe rain storm and also on ac-
count of lack of time. The docks and port facilities were in-
spected during a river trip tendered by the Board of Commis-
sioners of the Port of New Orleans.
On Tuesday evening the party left New Orleans by a special
train for the visits to the sulfur, salt, and sugar region. The
first stop was made at Lake Charles where the train was met
by members of the Chamber of Commerce. After a compli-
mentary breakfast, the party visited the mines of the Union
Sulphur Co., where the entire process of sulfur recovery was
shown, including the drilling of the well and inspection of the
sulfur bearing limestone. The party watched with greatest
interest the stream of molten sulfur coming direct from one of
the wells, as well as the centrifugal pumps and pipe lines by
which the molten sulfur was transported.
The next stop of the Institute Special was at New Iberia
where the train was backed out to the salt mines. After being
lowered 525 ft. in the mine elevator the party had the unique
experience of standing in chambers some 50 to 60 ft. high and
fully as wide, hewn out of a solid block of salt several thousand
feet thick and nearly a mile square. Any doubts as to the
purity of the glistening crystals were removed by an examination
of the clear, transparent samples to be found almost at random
in the mine.
On Thursday morning a stop was made at Franklin. After
a complimentary breakfast the party was taken by autos to the
Stirling sugar factory which was producing raw sugar from sugar-
cane. This is one of the largest cane sugar factories in Louisiana,
having a capacity of 1900 tons of cane daily. After seeing this
factory the near-by cane fields were visited where the gathering
and transportation of the cane was in progress. Most of the cane
was transported from the field to the factory in wagons, as
numerous small sugar factories are located in this region.
From Franklin the Institute Special returned to New Orleans,
and at 7 : 40 p. m. Thursday the party left New Orleans for
Chattanooga, Tenn., where arrangements had been made for
visits to Wilson & Co., a by-product coke plant and a ferro-
silicon plant, as well as a trip to Lookout and Signal Mountains.
The train was 6 hrs.' late, and therefore the Chattanooga pro-
gram was canceled.
At the next stop at Roanoke, Va., the blast furnaces of the
Virginia Iron, Coal and Coke Company were visited, as well
as a near-by pyrites plant where pyrites cinder is treated with
acid to remove the copper and sulfur, then sintered and sent
to the blast furnace for the production of pig iron.
At 5 : 45 P. m. a stop was made at Luray, Va., where the
last visit of the meeting was made to the wonderful caverns of
Luray. The natural statuary, convoluted stalactites and music
produced from the stalactites were quite as interesting as the
scientific aspects of these magnificent calcareous formations.
During the business sessions at New Orleans, resolutions
were adopted and wired to Washington urging the passage of
the Nolan bill, without the rider authorizing the exploitation
of patents by government employees, also the passage of the
Longworth dye bill.
President David Wesson was reelected for another year, as
were the secretary, John C. Olsen, the treasurer, F. W. Frerichs,
and the auditor, Chas. F. McKeuna. In the place of the three
retiring directors, F. M. de Beers, A. C. Langmuir, and T. B.
Wagner, there were elected F. E. Dodge, A. H. Hooker, and
Wm. D. Richardson.
The membership of the Society is now 454, the net increase
for the year being 89.
The attendance at the meeting was excellent both by out-of-
town members and by the local chemists and chemical engi-
neers. The meeting as a whole was very successful and en-
joyable, particularly on account of the generous hospitality ex-
tended at every place visited.
Brooklyn Polytechnic Institote J- C. OLSEN, Secretary
Brooklyn, N. Y.
ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS
The Thirty-Seventh Annual Convention of the Association of
Official Agricultural Chemists was held at the New Willard
Hotel, Washington, D. C, November 15 to 17, 1920. Over 300
members and visitors were present.
The usual reports of referees, associate referees, and com-
mittees were presented, and a number of special papers were
read. Interesting papers on the determination of borax in
fertilizers were presented. Papers on the present official method
and on a proposed method for insoluble phosphoric acid in
dicalcium phosphate resulted in lengthy discussion in which
many members participated. A paper dealing with the prep-
aration of neutral ammonium citrate was of special importance.
Honorable Edwin T. Meredith, Secretary of Agriculture,
spoke a few words of encouragement. Addresses were de-
livered by the president. Dr. H. C. Lythgoe, State Board of
Health, Boston, Mass., on "The Application of the
Theory of Probability to the Interpretation of Milk Analyses,"
and by the honorary president, Dr. Harvey W. Wiley, Wash-
ington, D. C, on "The Importance and Value of Agricultural
Research."
The following committee was appointed to cooperate with the
American Society for Testing Materials in the preparation of
specifications and testing for lime: W. H. Mclntire, Agri-
cultural Experiment Station, Knoxville, Tenn., chairman; Wm.
Frear, State College, Pa. ; and F. P. Veitch, Bureau of Chemistry,
Washington, D. C.
The following officers were appointed for the ensuing year:
President: W. F. Hand, Agricultural Collegt, Agricultural College,
Miss.
Vice President: F. P. Veitch, Bureau of Chemistry, Washington, D. C.
Secretary-Treasurer: C. L. Alsberg, Bureau of Chemistry, Wajk-
ington, D. C.
Additional members of the Executive Committee are:
A. J. Patten, Agricultural Experiment Station, East Lansing, Mich.
H. D. Haskins, Agricultural Experiment Station, Amherst, Mali.
The names of members of committees and of referees appointed
may be secured through the secretary, C. L. Alsberg, Bureau of
Chemistry, Washington, D. C.
CALENDAR OF MEETINGS
American Ceramic Society — Annual Meeting, Deschler Hotel,
Columbus, Ohio, February 21 to 24, 1921.
American Electrochemical Society — Spring Meeting, Hotel
Chalfonte, Atlantic City, N. J., April 21 to 23, 1921.
American Chemical Society — Sixty-first Meeting, Rochester,
N. Y., April 26 to 29, 192 1.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
91
PERKEN MEDAL AWARD
Announcement is made by the Committee of Award that the
Perkin Medal for 192 1 has been awarded by the American Sec-
tion of the Society of Chemical Industry to Dr. Willis R.
Whitney, Research Director of the General Electric Company,
in recognition of his distinguished work in the chemical field.
The presentation of the medal to Dr. Whitney will be made
at the regular meeting of the American Section of the Society of
Chemical Industry, in Rumford Hall, Chemists' Club, New
York, N. Y., on January 14, 1921.
CORPORATION MEMBERS OF THE AM ERICAN CHEMICAL SOCIETY
Abbott Laboratories Co., The
Amalgamated Dyestuff & Chemical Works. Iue.
Agricultural Chemical Co.
1 Cellulose & Chemical Mfg. Co . Ltd.
C li.<
1 Co., In
American Optical Co.
American Trona Corporation
American Zinc, Lead & Smelting Co.
Anaconda Copper Mining Co.
Antiseptol Liquid Soap Co. t
Arbuckle Brothers
Arkell Safety Bag Co.
Arlington Mills
Armour Glue Works
Arnold Print Works
Baker, H. J. & Bro.
Barrett Co., The
Bausch & Lomb Optical Co.
Beaver Board Companies, The
Binalbagan Estate, Inc.
Bishop & Co., J., Platinum Works
Bour Refractories Co., L. J., Inc.
Braender Rubber & Tire Co.
Brown Co., The
Bush & Co., W. J., Inc.
Calco Chemical Co.
California & Hawaiian Sugar Refining Co.
Cambridge Color & Chemical Co.
Carnotite Reduction Co.
Chemical Catalog Co., Inc.
Chemical Company of America, Inc.
Coal Tar Products, Inc.
Coca Cola Co.
Colgate & Co.
Commonwealth Chemical Corporation
Campagnie National de Matieres
Colorantes & de Produits Chimiques
Compagnie des Forges de Chatillon Commentry
et Neuves-Maisons
Consolidation Coal Co.
Contact Process Co.
Davison Chemical Co., The
Dearborn Chemical Co.
Diamond Alkali Co.
Dow Chemical Co
Drakenfeld & Co., B. F., Inc.
Drying Systems, Inc.,
Eastern Malleable Iron Co.
Electric Heating Apparatus Co.
Electro Bleaching Gas Co.
Eli Lilly & Co., The
Everlasting Valve Co.
Fairbank Co., N. K., The
Falls Manufacturing Co., The
Fels & Co.
Fisk Rubber Co., The
Garrigue & Co., William, Inc.
General Briquetting Co.
General Chemical Co.
General Tire & Rubber Co.
Gillette Rubber Co.
Gleason-Tiebout Glass Co.
Glidden Varnish Co.
Globe Soap Co., The
Grasselli Chemical Co.
Great Atlantic & Pacific Tea Co.
Great Western Sugar Co.
Hamilton & Sons, W. C.
Hammermill Paper Co.
Heath & Milligan Mfg. Co.
Heinze Co., H. J.
Herrick-Voigt Chemical Corporation
Heyden Chemical Works
Hommel Co., O.. The
Horween Leather Co.
Humboldt Mfg. Co.
Imperial Varnish & Color Co., Ltd., The
India Refining Co.
Interocean Oil Co.
Jeffrey Mfg. Co., The
Kelly-Springfield Tire Co
Kendall Mfg. Co.
Kewaunee Mfg. Co.
Kidde & Co., Walter, Inc.
Kimble Glass Co.
Kirk & Co., James S.
Kistler, Lesh & Co.
Knight, Maurice A.
Koppers Co., The
Krebs Pigment & Chemical Co., The
Lennig & Co., Charles
Lindsay Light Co.
Little, Inc., Arthur D.
Mallinckrodt Chemical Works
Merck & Co.
Merrell Co., Wm. S., The
Metal & Thermit Corporation
Midland Linseed Products Co.
Miehle Printing Press & Mfg. Co.
Milwaukee Coke & Gas Co.
Minnesota & Ontario Power Co.
Miranda Sugar Co.
Moorman Mfg. Co.
Morrill & Co., Geo. H.
Morris & Co.
Muralo Co.
National Aniline & Chemical Co., Inc.
Natural Products Refining Co.
New Jersey Zinc Co.
Newport Co., The
Niagara Alkali Co.
Nichols Copper Co.,
Norwich Pharmacal Co.
Noyes Bros. & Cutler, Inc.
Oakland Chemical Co.
O'Brien Varnish Co.
Onyx Oil & Chemical Co.
Patent Cereals Co.
Pennsylvania Rubber Co.
Peoples Gas Light & Coke Co.
Peterson & Co., Leonard, Inc.
Pfaudler Co., The
Philadelphia Quartz Co.
Pittsburgh Plate Glass Co.
Powers- Weigh tman-Rosengarten Co.
Procter & Gamble Co., The
Providence Dyeing, Bleaching & Calendering Co.
Rahr Sons Co., William
Raymond Bros. Impact Pulverizer Co.
Republic Chemical Co., Inc.
Riordon Pulp & Paper Co., Ltd.
Riverside Acid Works
Robeson Process Co.
Roessler & Hasslacher Chemical Co
Rohm & Haas
Rome Soap Mfg. Co.
Royal Crown Soaps, Ltd., The
Schoenhofen Co.
Sears, Roebuck & Co.
Sharpies Specialty Co., The
Shell Company of California
Sherwin-Williams Co., The
Singer Mfg. Co., The
Society Anonyme de Produits Chimiques de
Droogenbosch
Solvay Process Co.
Southern Cotton Oil Co.
Sowers Mfg. Co.
Special Chemicals Co.
Squibb & Sons, E. R.
Standard Parts Co.
Standard Ultramarine Co., The
Stanley, John T.
Steel Brothers & Co., Ltd.
Steere Engineering Co.
Swan Mfg. Co.
Swift & Co.
Talbot Dyewood & Chemical Co.
Tar Products Corporation
Thomas Co., Arthur H.
Thorkildsen- Mather Co.
Titanium Pigment Co., Inc.
Union Carbide & Carbon Corporation
Union Oil Company of California
United States Rubber Co.
Universal Oil Products Co.
Universal Portland Cement Co.
Valentine & Co.
Vanadium Corporation of America
Vulcan Detinning Co.
Wallace & Tiernan Co., Inc.
Welsbach Co.
Western Paper Makers Chemical Co
Whitall Tatum Co.
White Tar Co.
Whitmore Mfg. Co.
Will Corporation, The
Will & Baumer Co , The
Winkler & Bro. Co., Isaac, The
Wisconsin Steel Works
NOTES AND CORRESPONDENCE
PURE PHTHALIC ANHYDRIDE
Editor of the Journal of Industrial and Engineering Chemistry:
A United States patent1 has been granted to C. A. Andrews,
which claims as an article of manufacture "phthalic anhydride
in the form of colorless, needle-like crystals substantially chem-
ically pure and having a melting point above 130° C, corrected."
' U. S. Patent 1,336,182; filed Oct. 14, 1919; granted April 6, 1920.
In a recent article by H. D. Gibbs1 the fallacy of this claim has
been shown by reference to previous publications in chemical and
patent literature.
We are in position to substantiate Gibbs' statement with some
additional evidence. Pure phthalic anhydride in the form of
colorless, needle-like crystals and having a melting point above
1300 C. has not only been prepared previously in various labora-
1 This Journal, 12 (1920), 1017.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13. No. 1
tories but has been for many years a product of regular manu-
facture. It is true that organic handbooks, etc., give the melt-
ing point of phthalic anhydride as 1280 C, but it has been
known for some time by makers and users of this product that
the figure given in the chemical reference literature is about 3°
too low.
Prior to 1915 we imported phthalic anhydride during 6 or
7 yrs. from German and Austrian sources, and our analytical
records show that this product usually was of a very high de-
gree of purity and quite often had a melting point above 1300 C.
The melting point was determined on an average sample of
each shipment in the usual manner. In some cases the crystal-
lizing or solidification point was determined with 100 g. of the
product representing a composite sample from each of the
barrels of a shipment, and this crystallizing point also was
frequently found to be above 130° C. Comparative tests have
shown the melting point determined in a capillary tube to be
at least 0.5 ° higher than the crystallizing point determined as
described above.
Cryst. Pt. on
100-G. Sample
Date Bbls. ° C. Appearance
4/14/13 27 129.7 Short needles
5/16/13 4 130.3 Colorless needles
7/15/13 10 130.3 Colorless needles
7/14/14 10 130.7 Colorless needles
Melting Point in
Capillary Tube
4/1/14 8 131.5 Colorless needles
fi/20/14 4 130.5 Colorless needles
9/11/14 1 130-131 Colorless needles
The quality of the products of our own manufacture furnishes
additional evidence for the correctness of our contention. Prior
to the filing date of the Andrews patent we produced quantities
of phthalic anhydride in regular manufacture with a melting
point above 13 1° C, as shown by the following data taken from
our analytical records:
Cryst. Pt. on
100-G. Sample
Date Lbs. • C. Appearance
7/1/19 154 131.0 Colorless needles
7/8/19 367 131.1 Colorless needles
7/28/19 300 131.0 Colorless needles
8/14/19 175 131.0 Colorless needles
8/20/19 475 131.0 Colorless needles
9/5/19 400 131.0 Colorless needles
9/22/19 1405 131.0 Colorless needles
9/30/19 1075 131.0 Colorless needles
10/4/19 700 131.0 Colorless needles
In view of these facts it is evident that phthalic anhydride
having a melting point above 130° C. is not a new product and,
therefore, not patentable.
Monsanto Chemical Works JULES P.EBIE
St. Louis, Missouri
November 2a, 1920
STANDARDIZATION OF INDUSTRIAL LABORATORY
APPARATUS
Through the efforts of certain apparatus manufacturers, there
met informally at the Chemists' Club, New York City, on August
2, representatives of the following companies to discuss the
advisability of drawing up standard specifications for laboratory
apparatus to be used in their industrial research and works
control laboratories: Barrett Company, General Chemical
Company, Atmospheric Nitrogen Corporation, Grasselli Chemi-
cal Company, National Aniline & Chemical Company, New
Jersey Zinc Company, Solvay Process Company, Standard
Oil Company of New Jersey, and E. I. du Pont de Nemours
& Company.
Since most of these companies are members of the Manufac-
turing Chemists' Association of the United States, a committee
composed of these members was appointed by the Association
to pass on the proposals of the informal committee and to
recommend the adoption of the specifications resulting from the
informal committee's work as standard for the members of the
Manufacturing Chemists' Association.
Arrangements have been made for full cooperation with the
Committee on Guaranteed Reagents and Standard Apparatus
of the American Chemical Society, and also with the Committee
on Standards of the Association of Scientific Apparatus Makers
of the United States of America. These specifications will be
considered carefully by committees of these three societies, and
it is expected that they will then be published as tentative for
a period of 6 mo. in order to give time for general criticism.
At the end of that time the specifications will be adopted as
final.
In carrying on this work an effort will be made to obtain speci-
fications which will insure the cheapest mode of manufacture
of a given instrument consistent with the duties that it must
perform.
The committee desires to cooperate fully with all industries,
and any communications should be forwarded to the chairman,
Dr. E. C. Lathrop, E. I. du Pont de Nemours & Co.,
Wilmington, Delaware.
AMERICAN INSTITUTE OF BAKING, RESEARCH
FELLOWSHIPS
Arrangements have recently been made by the American
Institute of Baking by which the work done by its research
fellows at the University of Minnesota may be applied toward
the doctor's degree at that institution.
THE NOLAN BILL
Relief for the U. S. Patent Office, .although long delayed, is
apparently a prospect of the near future. The House has sent
the Nolan Patent Office reorganization bill to conference. The
bill was passed by the House last session and sent to the Senate.
There, during the closing hours of the session, Senator Norris of
Nebraska, chairman of the Senate Committee on Patents, was
forced to accept amendments so vitally changing the bill as
passed by the House that if enacted into law the result would
be a reduction in even the present force of the Patent Office.
The amendments were accepted, however, in order to assure
passage by the Senate during the last session, thus advancing
its parliamentary status.
Representative Nolan of California, chairman of the House
Committee on Patents, succeeded in having a special rule pro-
viding for sending the measure to conference between the House
and Senate by the end of the first week of the present session.
That all members of Congress are not supporters of the measure
is indicated by the opposition expressed on the floor of the
House. Representative Black of Texas made an effort to have
the House concur in the Senate amendments. The effect of
this would ba to enact the bill into law in the shape it passed
the Senate. This motion, however, was snowed under by a
vote of 210 to 154, and the measure sent to conference with
the House disagreeing to the Senate amendments
Representatives Nolan of California, Lampert of Wisconsin,
ranking Republican of the House Patents Committee, and Davis
of Tennessee, Democrat, were named as the House conferees,
while Senators Norris of Nebraska and Brandegee of Connec-
ticut, Republicans, and Senator Kirby of Arkansas, Democrat,
wort' named Senate conferees.
Attached to the Patent Office reorganization bill proper as
one of the Senate amendments is the measure providing for
acceptance and administration by the Federal Trade Commis-
sion of patents worked out by government scientists and tech-
nical experts. Senate conferees are desirous of keeping this
provision in the bill. House members, however, anxious that
the situation in which the Patent Office now finds itself be re-
lieved, fear that inclusion of this provision may be the cause
of the defeat of the entire bill, and will make a fight in confer-
ence to have it stricken out. Senator Norris is in favor of hav-
ing the provision remain in the bill. Other Senate conferees
also feel that the provision should be retained, and it is on this
question that the principal fight will ensue. There is no dis-
Jan., 1921 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
position on the part of the Senate conferees, Senator Norris said,
to insist on the Senate amendments reducing the salaries and
the working force of the Patent Office as provided in the bill
passed by the House. It is certain that they will readily agree
to increases both in the number of employees and remuneration
provided.
Meetings of the conferees are not expected much before
Christmas, and it is probable that no report will be made by
the committee until after the Christmas holidays. Final action
on a measure that will benefit the Patent Office within the near
future seems assured, however, both Senator Norris and Repre-
sentative Nolan being determined to press the measure.
the dye bill
Congress has swung into the second week of the third and
last session of the 66th Congress, and the fate of the dye bill is
still in question. The hearty promises that it would be imme-
diately pressed for action have begun to appear to even the most
hopeful of its supporters like the far-famed mirages of the
desert. At the present time there appears little probability
that action will be taken on either the dye or the several other
tariff measures pending in the Senate. Perhaps the most in-
teresting development in the dye situation has been the recent
frankness of Senator Moses of New Hampshire, who has waged
such a determined fight against the licensing feature of the bill
"because it violates principles I espouse."
Upon his return to Washington prior to the opening of this
session, Senator Watson of Indiana, in charge of the bill, declared
his intention of pressing the bill for action. He deemed it im-
possible, he said, to secure enactment of the measure with the
licensing feature embodied in it and consequently intended to
abandon that in favor of a system of tariff protection for the indus-
try. During a recent meeting of the Senate Finance Committee
the various tariff bills were discussed. Senator Watson said that
he did not think it would be possible to obtain passage of the dye
bill if it was to be amended by tacking on other tariff legisla-
tion for the purpose of using the dye bill as the vehicle to carry
through measures which otherwise would not be acted upon.
Senator Thomas of Colorado, Democrat, who distinguished him-
self last session by occupying the Senate floor for a week in
filibuster against the bill, said that he saw no reason why the
dye bill should not be amended so as to include the tungsten,
magnesite, "and in fact all the other tariff measures we have
here."
Senator Moses heretofore has been emphatic in his declara-
tion that his opposition was solely to the licensing feature of
the bill. The Senator possibly still holds that position. Never-
theless, in the face of declarations by Senator Watson that he
would abandon the licensing provision in favor of tariff protec-
tion, the New Hampshire Senator declared that if the dye bill
was to be acted on at this session he saw no reason why he
should not propose several amendments himself affording pro-
tection to textile machinery. This attitude of Senator Moses
can hardly be explained in view of his previous declaration.
WOOD CHEMICAL INDUSTRY CONFERENCES
The general business depression now existing, the lack of an
export market, and competition from Canada are the outstand-
ing problems facing the American wood chemical industry. Dis-
cussions at conferences held by the U. S. Tariff Commission
in Detroit December 7, and in Buffalo December 9 and 10,
■ 1920, with manufacturers, including representatives of the
Canadian industry, centered upon these obstacles. The Com-
mission was represented at these hearings by Commissioner
Edward P. Costigan and C. R. DeLong of the staff of chemical
experts of the Commission. Eight manufacturers were present
at the meeting in Detroit. At Buffalo the commission repre-
sentatives went over the situation at a conference with approx-
imately fifty, domestic manufacturers, on December 9, attending
a meeting of the National Wood Chemical Association. Two
Canadian representatives of the wood-distillation industry con-
ferred with Commissioner Costigan and Mr. DeLong the fol-
lowing day. One of . these represented the Canadian Electro
Products Company of Shawinigan Falls, Quebec, manufacturers
of synthetic acetic acid. Cooperation with the Commission in
its efforts to ascertain pertinent facts is understood to have
been promised by the Canadians.
The general business depression which now holds the business
of the nation for the most part in its grip, the decline — perhaps
to be expected to some extent — in the foreign sales, and the
competition that is being felt from the production in Canada of
synthetic acetic acid have left most American manufacturers
discouraged and depressed.
GERMAN COMPETITION IN THE DYE INDUSTRY
Congress and perhaps the country generally, inclined to dis-
count as extravagant the pictures of the probable competition to
be expected from Germany's dye trust painted by the proponents
93
of adequate protection for the American industry, is having the
enormous power of that country impressed upon it by the repre-
sentatives of many other American industries. Testifying before
the House Ways and Means Committee, urging adoption of
legislation that would equalize foreign exchange for the purpose
of assessing import duties, Franklin W. Hobbs, president of the
Arlington Mills, told the committee that "in dyestuffs for in-
stance, unless something is done we will be unable to meet the
competition and there will be no business left in this country.
Our industries will be wiped out." Mr. Hobbs was speaking in
favor of enactment of legislation that would protect the wool
manufacturer.
While perhaps there may be little to cause excitement in the '
mere announcement appearing recently in press dispatches from
Germany of the intention to establish in the United States and
m England German plants for the production of nitrate, advo-
cates of an American dye industry are inclined to see beneath
the surface the entering wedge of dangerous competition. It
is important to know whether the plant which it is proposed to
establish in this country will make ammonia or ammonium
sulfate, used for fertilizers, or go a step farther and produce
nitric acid, thus opening the way to the manufacture of aniline
and dye intermediates. It is significant that it is proposed to
establish such plants only in England and in the United States.
While our dye industry has, according to the best information
available, outstripped the development of the British industry,
these two promise the two sources of real competition to the
German industry. With Germany's past history of commercial
penetration in mind, one is inclined to view askance this newest
development and wonder if it is not another example of German
efficiency preparing to forestall the enactment of legislation ade-
quately protecting our industry and its proper development.
TARIFF REVISION
Desirous of having the new Republican revision of the tariff
on the statute books as soon as possible, the House Ways and
Means Committee has decided to begin tariff hearings on gen-
eral revision January 5. The Committee plans to go through
the present law schedules in alphabetical order, and on that
date proposes to take up Schedules A dealing with chemicals.
FOREIGN TRADE STATISTICS
Enlarged detail of import and export statistics, which has been
planned by the Bureau of Foreign and Domestic Commerce of
the Department of Commerce to be put into effect January 1,
may be delayed because of the failure of Congress to grant the
funds necessary. Plans worked out some time ago provide for
a very great extension of the import and export classifications
now contained in published foreign trade statistics. At the
present time these statistics are compiled by the customs divi-
sion of the Treasury at the various ports of entry and exit,
and the totals each month are forwarded here for publication
by the Bureau of Foreign and Domestic Commerce. In order
to simplify and coordinate the work of compilation, collection,
and publication of the statistics, it is proposed to transfer the
entire task to the Bureau of Foreign and Domestic Commerce.
This plan has met with the approval of both the Secretary of
the Treasury and the Secretary of Commerce.
In response to the numerous demands from the business in-
terests of the country, the Bureau of Foreign and Domestic
Commerce has prepared new classifications which it had hoped
to put in effect on January 1, coincident with the change from
the fiscal to the calendar year basis of publication of statistics.
It is estimated that this work will require $400,000 annually,
and provision for this sum is made in the estimates for the special
urgent deficiency bill now before the House Appropriations Com-
mittee. Whether or not the plan will go through will depend
upon Congress. The appropriations requested, it is to be re-
membered, are not in addition to funds already used, but include
funds now used by the Commerce and Treasury departments
separately for the carrying on of their parts of the work which
it is proposed to coordinate.
Hearings are expected to be held sometime within the next
2 wks. Officials of the Bureau of Foreign and Domistic
Commerce are anxious to put into effect the new schedules with
the beginning of the new year, and if a favorable report is made
by the House Appropriations Committee they will consider that
it is the intention of Congress to grant the funds necessary,
and proceed. It will be necessary, however, that Congress take
affirmative action before the last 2 wks. of January, as other-
wise it will be impossible to put the new classifications into effect
for that month.
The chemical industries are particularly interested in these
new classifications, inasmuch as they involve considerable ex-
tension of detailed figures as to imports and exports of dyes and
other chemicals.
December 14, 1920
94
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
PARI5 LETTER
By Charles Lormand, 4 Avenue de l'Observatoire, Paris. France
As I told you in my preceding letter, petroleum researches
in France are being actively pushed, and certain districts, where
it is thought petroleum will be found, remind me, in their ani-
mation, of those of Fort Worth and Dallas, which I visited at
the beginning of 1919-
Up to the present time the only positive result obtained is the
boring of Puy de Crouelle, 5 kilometers from Clermont-Ferrand.
For a long time this district of Limagne has been considered
by French geologists as likely to contain petroleum; and in 191 8
Dr. Hamor, chief of the Petroleum Division of the U. S. Bureau
of Mines, told me he thought petroleum researches should be
pursued in that district. His predictions were right, since the
present boring yields oil. The trial boring, which is 550 meters
deep, yielded 25 bbls. This oil is rather heavy, and contains
a rather high percentage of sulfur compounds (hydrogen sul-
fide and sulfur). It is supposed that it comes from the top part
of the deposit and that this part is somewhat oxidized, but that
at a greater depth lighter oils will be obtained.
The borings are made for the French government, and also,
in this same region by a Franco-Belgian company. Professor
Glangeaud, of the Faculty of Sciences of Clermont, is in charge
of the geological side of the work.
Another layer has also been reported in the Landes Depart-
ment, west from Saint-Sever. In that district the boring is
far less advanced, although geologists think that this layer
extends to the Lower Pyrenees. There also exist in this dis-
trict bituminous layers which are already being exploited.
Finally another layer is reported in the Alpine distiict, be-
tween the Rhone valley and that of the little river, Le Feir.
At the opening meeting of the Societe de Chimie Industrielle
on November 25, Professor Gentil, of the Faculty of Sciences
of the University of Paris, gave a summary of the present state
of geological information on petroleum prospecting. The contro-
versies between partisans of the mineral volcanic theory and
those of the organic theory are violent for, according to the point
of view, prospecting may be directed along very different lines.
A partial state monopoly is considered, but that project does
not seem to have great chance of succeeding, as the majority
of Parliament stands strongly against it.
THE DYESTUFF SITUATION
We are beginning to derive benefit from our efforts, made dur-
ing the war and since the armistice, not to be tributary to Ger-
many as regards dyestuff materials.
The "Compagnie Nationale des Matieres Colorantes" and the
"Soci£t6 des Produits Chimiques et Colorants Francais" were
amalgamated at the beginning of this year. These two companies
control about 70 per cent of the production, the remainder
being controlled by the "Societe de Saint-Denis," the "Societe
Alsacienne de Produits Chimiques de Thann et Mulhouse,"
the "Compagnie Francaise de Produits Chimiques et de
Matures Colorantes du Rhone," etc. German companies which
had factories in France are working under sequestration and
under the management of the "Compagnie Nationale."
The total output of all the manufactures, during the war and the
initial period, was 100 tons, jumped to 176 in June 1919, to 470
tons in January 1920, and finally to 764 tons in August 1920.
The monthly capacity of the French market is about 1000 tons.
The coloring materials we are lacking are specially alizarins,
certain basic dyes, and vat dyes.
The manufacture of intermediates has been partly ensured
by the transformation of munition factories.
INTERNATIONAL PATENTS
The French government has just agreed to the international
arrangement for the creation, in Belgium, of a central bureau of
patents. About 12 other nations have also agreed.
This Bureau, set up in Brussels, is to be an organ of docu-
mentation and of centralization as regards patents, from both
the legal and technical point of view. It has charge of the
international registration of applications for patents, and of
the transmission to the administrations of the adhering coun-
tries of applications for patents in one or several countries.
Furthermore, it will examine the applications and will proceed
to the necessary investigations regarding priorities.
Mr. J. C. Pennie's suggestions, made at the International
Chemical Conference in 1919, have been taken into considera-
tion. This is the first step towards the creation of an interna-
tional patent, which, although giving to the inventors the bene-
fit of legislation in their respective countries, will at the same
time safeguard their interests in foreign countries.
The French representative in Brussels is M. Drouet.
INDUSTRIAL CRISIS
The industrial crisis which I reported is becoming more and
more intense and the market of chemical products is under-
going a real crash. Little by little stocks are disappearing,
and in spite of the high price of certain raw materials tributary
to the rate of exchange, the drop in prices approaches 50 per
cent of those of 1919. A consequent general decrease in the
cost of living is expected.
"LA CHIMIE ET LA GUERRE"
M. Moureu, the president of the "Union Internationale de
Chimie," has just published a book, "La Chimie et la Guerre,"
which is a record of all services rendered by chemists and chemical
industries of all the allied nations. This little book covers
more than the limits of the French speaking public. Besides
indicating all that has been accomplished by chemists for the
war, it contains a great number of general ideas on the making
of chemists and the part played by chemistry in the life of
modern societies.
THE BASSET PROCESS
In one of my previous letters, I spoke about a new process
for the manufacture of steel — the "Basset process" for the
direct production of steel without using blast furnaces. This
process is more and more discussed, and it does not yet seem to
be out of the trial period. The big metallurgical firms look on
the process with reserve.
December 3, 1920
LONDON LETTER
By STBPBBN Miau., 28, Belsize Grove, Hampstead, N. W. 3, England
THE DYE BILL
Within the next few weeks Parliament must make a decision
as to the future of the dye industry in this country. Not only
is a great chemical industry essential to our future prosperity
but we cannot rely, as we have in the past, almost exclusively
on the manufacture of heavy chemicals, we must also have a
flourishing industry in the manufacture of aniline dyes, pharma-
ceuticals, and other synthetic organic compounds. The few
manufacturers of dyestuffs over here were occupied during the
war in the manufacture of poison gas and explosives, and toluene
was required for TNT rather than for toluidine; since the war
some progress has been made, but the present rate of the ex-
change between England and Germany enables Germany to
undersell the British manufacturers by a veiy considerable
margin. The government proposes to allow the German dye-
stuffs to be imported only by special license, and such license
would be refused when the British manufacturers can make the
dyestuff of good quality and sell it at a reasonable price. This
proposal, if carried, will give the British manufacturers the time
necessary for their gradual development, and though it will
be vigorously opposed by a number of the free traders over here,
it is generally expected that the government will be both wise
enough and strong enough to carry the measure through suc-
cessfully. By the time this letter reaches you the fate of the
bill will be pretty well known.
(The dye bill passed the House of Commons on December 18,
1920. — Editor)
the brunner, mond & company suit
We have been much interested in a law case recently. One
of the shareholders of Brunner, Mond & Co., Ltd., brought an
action to restrain the company from making a gift of £100,000
for educational purposes, on the ground that the company ought
not to spend its money except for its own benefit, and as the
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
95
proposed gift would necessarily benefit quite a lot of the other
people the proposal was ultra vires. The action was dismissed
by the judge, and I have not heard that the plaintiff has any
inclination to appeal. Had the decision been the other way,
the case would probably have been taken up to the House of
Lords, and if such a gift had there been held to be illegal there
was talk of introducing a bill in Parliament to make all such
gifts lawful. Indeed, the Federal Council for Pure and Applied
Chemistry had already sounded a few members of Parliament
to secure their assistance. An adverse decision would have been
a very serious blow to British chemistry, for our universities
cannot train sufficient chemists without such generous donations,
and private individuals in this country are, since the war, not
so comfortably situated as to find the necessary funds themselves.
CONFERENCE ON BRITISH PERIODICAL CHEMICAL LITERATURE
The Federal Council has within the last few days invited the
Chemical Society and the Society of Chemical Industry to ap-
point delegates to a joint conference on the periodical chemical
literature in this country. In many of the principal countries
this problem has already been successfully solved. America,
Holland, Italy, and some others come into one's mind, but in
France and Britain there is hardly any cooperation between the
societies who publish the transactions and abstracts of pure
chemistry and those who make public the new results of indus-
trial chemistry. In Britain the problem is both acute and com-
plex. Both the Societies are troubled by the high cost of print-
ing and paper and by exchequers largely depleted; each has its
own clientele, traditions, and staff; neither can afford to run
any risk of a reduced circulation and a corresponding loss of
revenue from advertisements, and the joint conference will have
to consider very carefully whether some species of cooperation
can be evolved which will effect economy in publication without
loss of revenue. Your experience in America, I am sure, will
be a valuable guide to the British committee, and if their delibera-
tions are not concluded before the summer we may learn a great
deal from you in a quiet talk round a bottle of any sustaining
fluid which the ingenuity of man may devise and procure for the
purpose. Water's the best of drinks, they say, and all the poets
sing, but who am I, that I should have the best of anything!
INTERNATIONAL LABOR ORGANIZATION
We are now seeing the first fruits of the International Labor
Conference which was held in Washington in November 1919.
This conference was presided over by a distinguished Ameiican,
but your country did not in any other respect take a conspicuous
part in the deliberations of that assembly. The conference dealt
with a variety of subjects, including diseases of occupation such
as anthrax and lead poisoning, and a recommendation was finally
adopted to prevent the employment of women and young per-
sons in processes likely to produce lead poisoning. Those of
us who attended the conference found we had plenty of work to
do, and the discussions were the more difficult in that they were
usually bilingual. ^When we came to translate the Washington
recommendation into the Act of Parliament we found it no easy
task to make the terms of the recommendation fit in with our
existing legislation and our special industrial conditions, and the
House of Lords has had to listen to details as to solubility of
lead compounds, the manufacture of lead silicates, and the de-
termination of lead in solution by precipitation and estimation
as lead monoxide. I believe all of us who have been through this
experience realize how much time and how much attention to
detail is necessary for the proper application of such general
ideas as may appear to be feasible, and how important it is
that the international labor organization shall consider such
highly technical matters as injurious processes in a detailed and
leisurely manner impossible in a hurried conference.
FUEL ECONOMY
Fuel economy has been before the public ever since I can
remember, and the number of schemes to enable us to save 10
or more per cent of our coal or money is almost infinite. The
advocates of high-temperature carbonization, of low-temperature
carbonization, of dry carbonizing and wet carbonizing have been
busy in the press and on the Stock Exchange. It is an extraor-
dinary thing that for power purposes nothing seems to be
cheaper than a well-conducted boiler of the old-fashioned type
heated by ordinary coal. Its elasticity and simplicity seem to
counterbalance and even more than counterbalance the waste
of benzene, toluene, ammonia, and phenol. Powdered fuel,
colloidal fuel, gas, and oil are still in an experimental stage. I do
not know whether all the permutations and combinations of
solid, liquid, and gaseous fuel have yet been investigated, but
a good many are still under discussion. After many years of
doubt and disaster I am now informed that low-temperature
carbonization has been got to work satisfactorily. The diffi-
culties in the past have been largely mechanical and seem to
have been surmounted. It seems that the new plant at Barnsley
in Yorkshire is working well and that there is a reasonable chance
that the patience of the shareholders will ultimately be justified.
All the metals seem to be having a race as to which can reach
the bottom first, and as no one cares to buy on a falling market
the trade in inorganic compounds is extremely limited. I
imagine that this phenomenon must be very prominent on your
side of the Atlantic as well as this, and it is hard to say whether
the outbreak of war or the outbreak of peace has been the more
disastrous.
The visit of the Society of Chemical Industry to Canada and
the United States next September already causes much interest
over here and the program, so far as it is known, is most attrac-
tive. In the future no nation can be a great industrial nation
unless it is a great chemical nation, and we have much to learn
from the well-organized chemical industries in these two coun-
tries and from the chemists whom too few of us know personally.
December 6, 1920
PERSONAL NOILS
Mr. Regis Chauvenet, president emeritus of the Colorado
School of Mines, chemist and metallurgist, died in Denver
recently at thejagejof seventy-eight.
Dr. Elijah P. Harris, emeritus professor of chemistry at
Amherst College, died recently at Warsaw, N. Y , at the age
of ..eighty-eight. Dr. Harris retired as professor of chemistry
at Amherst in 1907 and became emeritus professor on the Car-
negieJFoundation. He was the author of a book on "Qualita-
tive Analysis" which went through ten editions.
Mr. Harry W. Eberly, acid assistant in charge of nitric acid
at the Forcite Works of the Atlas Powder Co., Landing, N. J.,
and a member of the American Chemical Society, died last
October at the Dover General Hospital from the effect of nitric
acid fumes received from a spill in the nitric acid house of which
he was in charge.
Mr. Isaac Neuwirth is now associated with Dr. Israel S.
Kleiner, as instructor in physiological chemistry at the New York
Homeopathic Medical College and Flower Hospital, New York
City.
Mr. Sherman Leavitt, formerly with the Illinois State Water
Survey Division at the University of Illinois, has been appointed
instructor in food chemistry and technical analysis at the Uni-
versity of Minnesota, Minneapolis, Minn.
Mr. Robert A. Miller, Jr., formerly with the Stillwell & Glad-
ding Co., of New York, is at present engineering research chem-
ist with the Rubber Regenerating Co., of Naugatuck, Conn.
Mr. H. O. Bernstrom, until recently with the Lignol Chemical
Co., Irvington, N. J., where he was working on hardwood oils,
is now attached to the chemical and research division at Edge-
wood Arsenal, Edgewood, Md.
Mr. H. E. Brown, of New York City, has been appointed
engineer of the plant of the Bartholomay Co., Inc., at Rochester,
N. Y. This plant was formerly the Genesee Brewery, and the
Bartholomay Company has let a contract for converting it into
a vegetable oil refinery, using the Brown-Baskerville process.
Mr. Floyd E. Rowland, assistant professor of chemistry at the
University of Kansas last year, has been elected head of the
department of chemical engineering at the Oregon Agricultural
College, Corvallis, Ore.
Mr. E. G. Gross has resigned as instructor of agricultural
chemistry at the University of Wisconsin, and is holding a
fellowship in the Yale Graduate School in the department of
physiological chemistry with Dr. Mendel.
Mr. John Gore, formerly assistant superintendent of the
Russ Gelatin Co., Westfield, Mass., has become chemical en-
gineer for the Beech-Nut Packing Co., Canajoharie, N. Y.
Mr. C. G. Smith, who has been connected with the Dow
Chemical Co., Midland, Mich., for the past five years, in
the capacity of experimental chemist and engineer, resigned
last spring because of ill health, and is at the present time teach-
ing science in the Canon City High School, Canon City, Colo-
rado.
96
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
Drs. Frederic C. Lee and E. Hyatt Wight have formed a part-
nership under the firm name of Lee & Wight, and have opened
a consulting and analytical laboratory in Baltimore, Md.
Mr. Ellery L. Priest, formerly with the W. S. Merrell Chemical
Co., Cincinnati, Ohio, has joined the firm of the Western Chem-
ical Co., Hutchinson, Minn., where he is assistant chemist.
Mr. Charles A. Fort has left the General Electric Co., of
Pittsfield, Mass., where he was employed as research chemist
on insulating materials, and has become chief chemist for the
Forest Products Chemical Co., of Memphis, Tenn. His new
work consists mainly of research on hard-wood tar products.
Mr. Louis Mittelman resigned his position with the Sun Com-
pany at their Toledo refinery to accept a position as chemist
with the Associated Oil Co., at Gaviota, Cal.
Mr. Jesse E. Day severed his connections this past summer
as assistant professor of general chemistry at Ohio State Uni-
versity, Columbus, O., to become assistant professor of general
chemistrv for engineers at the University of Wisconsin, Madison,
Wis.
Mr. K. V. Froude, who for the last two years has been assis-
tant chemist in the laboratory of the Bettendorf Steel Works,
Bettendorf, Iowa, has been promoted to the position of chief
chemist_for the same company.
Mr. A. E. Plumb, until recently chief chemist for F. J. May-
wald, consulting rubber technologist of Newark, N. J., with
laboratory at Nutley, N. J., now holds a similar position with
Hodgman Rubber Co., of Tuckahoe, N. Y.
Dr. Irene C. Diner, previously attached to the division of in-
dustrial chemistry at New York University, New York City,
has become associated with the research division of the Chem-
ical Warfare Service, in the capacity of associate chemist work-
ing on rubber problems.
Dr. C. B. Clevenger resigned an instructorship in the depart-
ment of chemistry, University of Wisconsin, Madison, Wis.,
to accept a professorship of agricultural chemistry and head of
the department of chemistry of the Manitoba Agricultural Col-
lege, Winnipeg, Canada.
Mr. Isador W. Mendelsohn, chemist and state sanitary engi-
neer of the State Board of Health of North Dakota for the past
two years, has become assistant sanitary engineer of the Bureau
of the Public Health Service, detailed at Washington, D. C.
Mr. C. W. Leggett is at present employed by the McCall
Cotton & Oil Co., Phoenix, Ariz., as superintendent and
chemist.
Mr. C. K. Jones has resigned from the Van Camp Packing
Company in order to accept the position as chief chemist for
the Whitman Candy Co., Philadelphia, Pa.
Mr. A. E. Koenig, who was assistant professor of chemistry
at the University of Wisconsin, Madison, Wis., has resigned
from that position and is now at the State School of Mines,
Butte, Mont., as associate professor.
Mr. Joseph V. Meigs, formerly connected with the New
Jersey Testing Laboratories, Montclair, N. J., as research chem-
ist, is chief chemist for the Massachusetts Oil Refining Co., at
East Braintree, Mass.
Mr. Rolla N. Harger has resigned as assistant biochemist,
Soil Fertility Investigations, Bureau of Plant Industry, Wash-
ington, D. C., to accept one of the National Research Council
fellowships in chemistry. Mr. Harger's work will be on a prob-
lem in organic chemistry and will be done at Yale University,
New Haven, Conn.
Mr. R. H. Currie has left the du Pont Company of Wilming-
ton, Del., where he was attached to the main office chemical staff,
and is at present with the Acheson Graphite Co., Niagara
Falls, N. Y., as assistant superintendent.
Mr. Harold J. Barrett has been appointed instructor in chem-
istry at Iowa State College, having come there from West Vir-
ginia University, Morgantown, W. Va.
Mr. Phil G. Horton has recently resigned his position as chem-
ist in the research laboratory, film section, of E. I. du Pont de
Nemours & Co., Parlin, N. J., and is taking a postgraduate
course in chemistry at Ohio State University.
Mr. J. Irving Prest, formerly chemist at the Pacific-Northwest
Experiment Station of the U. S. Bureau of Mines, Seattle,
Wash., has joined the forces of the International Harvester Co.,
Chicago, 111.
Dr. S. A. Mahood, who has been in charge of investigations on
wood cellulose and essential oils at the U. S. Forest Products
Laboratory, Madison, Wis., for the past three years, has be-
come associate professor in charge of organic chemistry at Tu-
lane University, New Orleans, La.
Miss Mary V. Buell, who taught nutrition in the home eco-
nomics department of the University of Wisconsin, Madison,
Wis., last year, is at present teaching chemical dietetics and phys-
iological chemistry in the home economics department of the
University of Iowa, with headquarters at the University Hos-
pital of the State University of Iowa, and is also cooperating
with the medical staff in their metabolism work and research.
Dr. Ernest Anderson, for the past three years professor of
agricultural chemistry in the University of South Africa, has
been appointed professor of general chemistry in the University
of Nebraska, Lincoln, Neb.
Mr. Frank Bachmann resigned his position as chief chemist,
Industrial Waste Board, Connecticut State Department of
Health, to accept a position in the sanitary engineering depart-
ment of the Dorr Company of New York City.
Mr. Floyd A. Bosworth, formerly junior chemist in the United
States Food and Drug Inspection Station at Buffalo, N. Y., is
now employed in the research and analytical department of the
United Drug Company at Boston, Mass.
Mr. Henry Ward Banks, 3d, formerly research chemist with
the Harriman Laboratory and the National Biscuit Co., and
Mr. Robert Hall Craig, formerly with the office of the Surgeon
General of the Army, Washington, D. C, and later with the con-
struction division of the Army, have formed a partnership under
the name of Banks and Craig, consulting engineers and chem-
ists, in New York City. Dr. D. D. Jackson, of Columbia Uni-
versity, is associated with the firm in the capacity of consulting
sanitary engineer.
The following have become members of the staff of the de-
partment of chemistry of the College of the Citv of New York:
W. McG. Billing, H. P. Coats, Alexander Cohen, A. C. Glennie,
Nathan Hecht, and F. D. SneU.
Mr. C. B. Wiltrout, formerly chief chemist for the Continental
Sugar Co., Toledo, Ohio, has been engaged as chief chemist by
the raw sugar refining interests of the Independent Sugar Co.,
Marine City, Mich.
Mr. J. S. Staudt has become associate professor of electrical
engineering at Texas A. & M. College, College Station, Texas.
He was formerly in the government employ at the Old Hickory
Powder Plant near Nashville, Tenn.
Mr. Hugo H. Sommer has resigned as chemist for the Northern
California Milk Producers Association, Sacramento, Cal., to be-
come assistant professor of dairy husbandry in the dairy depart-
ment of the University of Wisconsin, Madison, Wis.
Dr. Frederick E. Breithut has entered the employ of the Calco
Chemical Company, Bound Brook, N. J.
Dr. William C. Moore, until recently associated with the
School of Hygiene and Public Health of Johns Hopkins Uni-
versity, is now on the research staff of the United States In-
dustrial Alcohol Co., Baltimore, Md.
Dr. Frederick W. Lane, for the past three years instructor
in chemistry at Yale University, has become organic chemist
in the petroleum division of the Pittsburgh Station, U. S. Bureau
of Mines, Pittsburgh, Pa.
Dr. M. E. Holmes, formerly research engineer for the Na-
tional Carbon Co., Cleveland, Ohio, has been appointed manager
of the chemical department of the National Lime Association,
Washington, D. C.
Mr. Bartholomew O'Brien, formerly with the Synfleur Scien-
tific Laboratories of Monticello, N. Y, has joined the staff of
the Grasselli Chemical Co., Albany, N. Y.
Mr. Kirby E. Jackson, head of the science department at the
Marion County High School, Jasper, Tenn., has been appointed
professor of chemistry at the Daniel Baker College, Brownwood,
Texas.
Miss Martha G. Barr, who was instructor in chemistry at
Iowa State College, Ames, Iowa, from 1918 to 1920, now has
charge of the chemical laboratory of the Lane Cotton Mills of
New Orleans, La.
A recent acquisition to the engineering staff of the John
Johnson Co., Brooklyn, N. Y., is announced in the per-
son of Capt. Wilkinson Stark, late of the Army Ordnance De-
partment. Prior to his service in the Army, Captain Stark was
employed by the du Pont Company, who released him at the
beginning of the war to supervise the design, installation, and
operation of the Army's caustic recovery and cotton purifica-
tion, bleaching, and drying divisions at Explosives Plant "C,"
Nitro, W. Va.
Dr. Edward Schramm, formerly research chemist with the
Bridgeport Brass Co., Bridgeport, Conn., is now with the
Onondaga Pottery Co., Syracuse, N. Y., as research chemist.
Jan., 1021
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
Q7
GOVERNMENT PUBLICATIONS
By Nellie A. Parkinson, Bureau of Chemistry, Washington, D. C.
NOTICE — Publications for which price is indicated can be
purchased from the Superintendent of Documents, Government
Printing Office, Washington, D. C. Other publications can
usually be supplied from the Bureau or Department from which
they originate. Commerce Reports are received by all large
libraries and may be consulted there, or single numbers can be
secured by application to the Bureau of Foreign and Domestic
Commerce, Department of Commerce, Washington. The regu-
lar subscription rate for these Commerce Reports mailed daily
is $2.50 per year, payable in advance, to the Superintendent of
Documents.
DEPARTMENT OF LABOR
Employment of Women in Hazardous Industries in the
United States. Summary of State and Federal Laws Regulating
the Employment of Women in Hazardous Occupations, 1919.
Bulletin 6 (Reprint). 8 pp. 1920.
NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS
Comparison of Alcogas Aviation Fuel with Export Aviation
Gasoline. V. R. Gage, S. W. Sparrow and D. R. Harper, 3d.
14 pp. Report 89. Paper, 5 cents. 1920.
Comparison of Hector Fuel with Export Aviation Gasoline.
H. C. Dickinson, V. R. Gage and S. W. Sparrow. Report 90.
10 pp Paper, 5 cents. 1920.
NAVY DEPARTMENT
Instructions for Care and Operation of Fuel Oil-Burning
Installations. Revised edition, 1920. 90 pp.
WAR DEPARTMENT
Aviation Gasoline, Specifications and Methods of Testing.
Prepared by Material Section of Air Service. Air Service
Information Circular, Heavier-than-Air, Vol. 1, No. 46, Aug.
30, 1920. 8 pp.
Report of Tests of Metals and Other Materials Made in
Ordnance Laboratory at Watertown Arsenal, Mass., Fiscal
Year 1918. War Department Document 901, 338 pp. Paper,
80 cents. (In many cases one side of the leaf only is paged,
the unnumbered side usually bearing illustrations, although
in some cases it is blank.)
BUREAU OF FOREIGN AND DOMESTIC COMMERCE
Hides and Leather in France. Norman Hertz. Special
Agents Series, No. 200. 159 pp. Paper, 20 cents. 1920. The book
includes an introduction, general survey of conditions, a descrip-
tion of market requirements for leather, the domestic tanning
industry, foreign trade in leather, customs tariff, leather mer-
chandising, foreign trade in hides and skins, domestic hides and
skins and tanning materials, and an appendix. The conclusion
is drawn that while American tanners cannot expect to continue
the volume of business in France that was transacted during the
war and immediately after, the outlook for continued sales
of many kinds of leather, especially upper leather, is very
good, provided American manufacturers keep constantly in mind
the fact that it is better to keep a customer satisfied than to
make a few large sales.
PUBLIC HEALTH SERVICE
An Outbreak of Botulism at St. Anthony's Hospital, Oakland,
Cal., in pctober 1920. Public Health Reports, 35, 2858-60.
There was a total of six cases, two of which could be considered
mild and four severe. Of these latter, three died. Unfortunately,
none of these cases was recognized as botulism until the third
day of illness, and therefore they were not immediately reported.
BUREAU OF MINES
Monthly Statement of Coal-Mine Fatalities in the United
States, August 1920. W. W. Adams. 8pp. Paper, 5 cents.
October 1920.
Monthly Statement of Coal-Mine Fatalities in the United
States, September 1920. W. W. Adams. 8 pp. Paper, 5 cents.
Norember 1920.
BUREAU OF STANDARDS
Sodium Oxalate as a Standard in Volumetric Analysis.
Circular 40, 3d ed. 13 pp. Paper, 5 cents. 1920. This
circular is not issued for the purpose of publishing any new
information or of entering into a critical discussion of volumetric
standards, but rather to give a resume of the work done at the
Bureau of Standards and elsewhere which has led to the selection
of the sodium oxalate as a primary standard. This third edition
has been revised with special reference to the methods employed
and the results obtained in the testing of the second preparation
of sodium oxalate which is now issued as Standard Sample No.
40a.
!■ Recommended Specification for Composite Thinner for Thinning
Semipaste Paints when the Use of Straight Linseed Oil Is
Not Justified. Prepared and Recommended by the United States
Interdepartmental Committee on Paint Specification Standard-
ization, September 27, 1920. Circular 102. 5 pp. Paper, 5
cents. Issued October 18, 1920. This specification covers a
composite thinner which contains in one liquid drying oil. drier,
and volatile thinner. General specifications are given, and
methods of sampling, laboratory examination, and the reagent
employed are described.
Recommended Specification for Spar Varnish. Prepared and
Recommended by the United States Interdepartmental Commit-
tee on Paint Specification Standardization, September 27, 1920.
Circular 103. 5 pp. Paper, 5 cents. Issued October 18, 1920.
The specification provides that the varnish shall be the best long oil
varnish, resistant to air, light, and water. The manufacturer
is given the wide latitude in the selection of raw materials and pro-
cesses of manufacture, so that he may produce a varnish of the
highest quality. It must, however, comply with certain require-
ments, which are outlined. Methods of sampling and a descrip-
tion of the laboratory examination are described.
Recommended Specification for Asphalt Varnish. Prepared
and Recommended by the United States Interdepartmental
Committee on Paint Specification Standardization, September
27, 1920. Circular 104. 6 pp. Paper, 5 cents. Issued October
18, 1 920. The varnish must be composed of a high grade of
asphalt fluxed and blended with properly treated drying oil
and thinned to the proper consistency with a volatile solvent.
It must be resistant to air, light, lubricating oil, water, and min-
eral acids of the concentration specified, and must meet certain
requirements, which are outlined. Methods of sampling and
laboratory examination are also described.
A Study of the Relation between the Brinell Hardness and the
Grain Size of the Annealed Carbon Steels. H. S. Rawdon
and Emilio Jimeno-Gil. Scientific Paper 397. 37 pp. Paper,
10 cents. 1920.
Sulfur in Petroleum Oils. C. E. Waters. Technologic
Paper 177. 26 pp. Paper, 5 cents. October 20, 1920. Short
accounts are given of theories concerning the origin of the
sulfur and sulfur compounds which are found in crude petroleum.
The forms of combination in which the element occurs, their
identification, and significance are briefly discussed. Tests for
the detection of sulfur are described, and the copper test is shown
to be one of great delicacy. Various methods that have been
used for the determination of sulfur in oils, and finally a new
procedure, are described. Data obtained by the analysis of
certain oils by the new and other methods are given.
DEPARTMENT OF AGRICULTURE
Milk Plant Equipment. Ernest Kelly and C. E. Clement
Department Bulletin 890. 42 pp. Paper, 15 cents. Issued
October 1920. This bulletin points out some of the more im-
portant economic and sanitary problems in the handling and
distribution of milk.
Manual of Design and Installation of Forest Service Water
Spray Dry Kiln. L. V. Teesdale. Department Bulletin
894. 47 pp. Paper, 10 cents. Issued October 18, 1920.
Describes a kiln in which the temperature, humidity, and circu-
lation can be regulated independently of the others.
Weight Variation of Package Goods. H. Runkel. Depart-
ment Bulletin 897. 20 pp. Issued November 15, 1920.
Fumigation of Citrus Plants with Hydrocyanic Acid: Condi-
tions Influencing Injury. R. S. Woglum. Department Bulle-
tin 907. 43 pp. Paper. 15 cents. Issued October 20, 1920.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
Toxicity of Barium Carbonate to Rats. E. W. Schwartze.
Department Bulletin 915. 11 pp. Paper, 5 cents. Issued
November 12, 1920.
Cooperative. Cane-Sirup Canning: Producing Sirup of Uni-
form Quality. * J. K. Dale. Department Circular 149. 19 pp.
Issued November 1920. At present the cane-sirup industry is
handicapped by a lack of uniformity in the sirup offered for sale
by the individual farmer. This condition may be remedied by
the adoption of new and ^improved methods of manufacture
and by cooperative canning.
* Report of the Chemist. C. L. Alsberg. 30 pp. Issued
December 1920. This publication is a report of the work of
the Bureau of Chemistry for the fiscal year ended June 30, 1920.
Articles from Journal of Agricultural Research
Investigations of the Germicidal Value of Some of the Chlorine
Disinfectants. F. W.'Tuaey. 20 (October 15, 1920), 85-110.
Studies in Mustard Seeds and Substitutes: I— Chinese
Colza {Brassica campestris chinoleifera Viehoever). Arno
Veshoever, J. F. Clevenger and C. O. Ewing. 20 (October
15, 1920), 111-15.
Study of Some Poultry Feed Mixtures with Reference to
Their Potential Acidity and Their Potential Alkalinity. B. F.
Kaupp and J. E. Ivey. 20 (October 15, 1920), 141-9.
The Influence of Cold in Stimulating the Growth of Plants.
F. V. CovellE. 20 (October 15, 1920), 151-60.
COMMERCE REPOETS— NOVEMBER 1020
The Government laboratory of Jamaica has been conducting
experiments for the production of pimento-leaf oil from pimento
leaves. Pimento leaves yield about 1.8 per cent of eugenol,
from which isoeugenol and vanillin can successfully be obtained.
If a market can be found, Jamaica can produce 100,000 lbs.
of pimento-leaf oil per annum from materials at present wasted.
(P. 500)
An important financial group, representing English, French,
and Rumanian interests, has purchased the control of one of
the great oil producing companies of Rumania. So far as Great
Britain is concerned, more than £2,000,000 are involved in the
matter. (P. 501)
A process for the manufacture of flax-straw waste on a commer-
cial scale has been developed in Argentina. The product of
this new process is reported to be equal or even superior in color,
elasticity, length of fiber, and resistance to fibers retted by the
old methods, which required many days' time, as compared with
less than half an hour by the new process. (Pp. 520-1) _
The outlook for the Swedish iron industry is unfavorable.
(P. 53i) . -_
A good market is reported for American laundry soap in Bul-
garia. The soap must contain fats to the extent of at least 70
per cent. (P. 541)
The German process for artificial wool has proved unsuccess-
ful, as it was impossible to put the wool into solution without
a resultant decomposition. The application for a patent has
been abandoned. (P. 549)
The United States at present furnishes very nearly all the
dyes used in the district for which Tientsin is the distributing
center, and if American manufacturers are willing to meet the
requirements of the trade they will be in the market perma-
nently. (P. 553)
The great milling wealth of the Kongo is being rapidly devel-
oped, and the production of gold, copper, and diamonds is con-
stantly increasing. The war acted as a great stimulus on the
copper-mining industry. (P. 556)
It is reported that detailed research is shortly to be under-
taken in India with a view to determining the practicability
of producing power alcohol on a commercial scale. Meanwhile,
Great Britain is trying to make possible the ready use of such
substitute fuel whenever.it becomes available in sufficient quan-
tity. (P. 57i)
The Finnish Government is erecting a superphosphate factory
in Kotka and a sulfuric acid factory in Vilmanstrand. It is
estimated that the production of the former will amount to 20,000
tons, which will be sufficient to satisfy all domestic requirements
and probably leave a small surplus for export. The products
of4the sulfuric acid factory will be used for the most part in the
manufacture of superphosphate. (P. 578)
Remarkable success has attended the manufacture of linseed
oil in South Australia. (P. 582)
Samples of flax-straw fiber and waste made from flax straw
from Argentina are available for examination at the Bureau of
Foreign and Domestic Commerce. (P. 592)
There is a shortage of brass and copper in Switzerland which
would appear to offer quite a market for American copper.
(P. 594)
Remarkable results are being obtained in Germany from the
manufacture of yarn from grasses, plants, leaves, etc. (Pp.
595-7)
A market for industrial drugs and chemicals is reported in
Argentina. Tabular statements are given showing the principal
chemical products used in Argentina, the typical industries
using such products, and a price list of one Argentine dealer in
chemicals. (Pp. 630-3)
A translation is given of a decree relative to the exploitation
of petroleum mines in Salvador. (P. 649)
A Japanese government oil monopoly is being proposed
largely in order to guarantee supplies for the navy. (P. 658)
British prohibition of the importation of synthetic dyestuffs,
except under license, is proposed in order to foster the domestic
industry. (P. 673)
A decrease of 30 per cent is reported in the production of
olive oil in the Malaga district for the season 1920-2 1 as compared
with 1919-20. (Pp. 678-9)
The Bureau of Foreign and Domestic Commerce has ready
for distribution a list of importers and dealers in paints and
varnishes in China. (P. 680)
Rubber estates in Java are reported to have had a satisfactory
first half year. (P. 693)
A market is reported in France for American leathers. Ger-
many is in no position to make deliveries, and the United King-
dom is said to have no advantages over American tanners.
(P. 696)
The rubber market in the Straits Settlements has been marked
by a steady decrease in price from $0.50 per pound in January
1920 to about $0.23 in September. (P. 708)
Statistics are given showing the quantities of coal-tar dyes and
intermediates imported into the United Kingdom during the
first nine months of the current year. Comparative figures are
also given for finished dyes not only for the current year but
for the same period in 19 13 and 19 19, and the value of these
imports, converted into American currency, is also given. (Pp.
71 i-3)
Fifteen years ago Malaya produced over 60 per cent of the
world's tin; to-day the figure stands at less than 40 per cent.
Although the percentage comparison of Malayan output with
the world's total has fallen, owing to greater production elsewhere,
the actual outturn has considerably increased. (P. 715)
About 10,000 tons of citrate of lime and about 300 metric
tons of citric acid are held in Italy. (P. 737)
A market for sodium and potassium is reported in Argentina.
Sodium, in various forms, is employed in practically every
industry, large and small. Neither hydrate, carbonate, nor
silicate of sodium are made in Argentina on a commercial scale.
(Pp. 745-7)
The discovery of extensive deposits of pyrites a short distance
from Prague, Czechoslovakia, has caused considerable stir in
the industrial circles of that republic. (P. 747)
An agreement has been reached whereby the production of
rubber will be curtailed 25 per cent until December 1921. (P.
766)
Sulfur ores in Mexico are now available for shipment to the
United States. (P. 772)
Paints and varnishes are required by the Peking-Hankow
Railway and bids are called for these materials at quarterly
intervals. (Pp. 776-7)
The manufacture of acids in Argentina is described, as wall
as the uses to which other chemicals are put. (Pp. 779-82)
A note from Manitoba is to the effect that crude oil from
Texas wells is to be imported and refined and distributed in
Western Canada. (P. 792)
The production of yacca gum in South Australia, its use,
chemical reactions and destinations of exports are described.
(Pp. 796-7)
A new paper pulp industry has come into existence in Argentina.
A species of bog grass called "paja brava" is the raw product
employed. This grass grows during the whole year and is so
abundant in the swampy places that it has been considered a
nuisance. (P. 799)
A list of importers and dealers in chemicals in Australia may
be obtained upon request of the Bureau of Foreign and Domestic
Commerce. (P. 800)
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
99
The great depressi^ in the Amsterdam rubber market still
continues. (P. 817)
The proposed petrc^im law in Peru sets forth the conditions
under which concessjons of petroleum land will be made, the
maximum term of cc.ntracts being placed at 75 yrs. (P. 819)
The Bureau of Foregn and Domestic Commerce has ready
for distribution a list i>f oil mills and exporters of vegetable
oils in India. (P. 8:>o)
An American comoanV has secured a "gusher" in Trinidad
which has a daily production of about 1000 bbls. This will
probably prove to be the best oil ever drilled in Trinidad. (P. 825)
A shortage of fuel oil is reported in Vancouver. (P. 835)
The Japanese chemical market is still unsteady, sales decreas-
ing while holdings are being readjusted. (P. 836)
Recent advices state that the Japanese dyestuff industry can-
not successfully compete with the American or German manufac-
tures, even with the new import duty of 35 per cent. (P. 851)
The leather situation in Palestine is reviewed. (Pp. 857-8)
The wood-pulp market in Finland has been steady, but while
the cellulose market was exceedingly brisk in the spring, export-
ers are somewhat pessimistic about the future. (Pp. 869-72)
Statistics are given showing the output of the government
oil reserves at Comodoro Rivadavia, Argentina, from 1907 to
1918, inclusive. (P. 873)
The Chinese summer indigo crop is reported to have been
normal, though it suffered somewhat from floods. (P. 874)
A large Australian company has under consideration the ex-
tension of its manufacturing processes to substances not pre-
viously made in Australia and from which the chief by-product
will be chlorine. (P. 884)
The Canadian starch and glucose industry is reviewed.
(Pp. 885-6)
The German factory of Adler & Oppenheimer, considered
the largest leather factory in Europe, has being sold to a group
of French and Alsatian interests, and it is intimated that special
attention will be given to exporting the products of the factory.
(P. 903)
The Argentine market for calcium carbide, chloride of lime,
glycerol, glucose, and cryolite, barium, copper, iron, and mag-
nesium sulfates is described. (Pp. 906-7)
New import duties in Peru are announced for the following
materials: chemicals, drugs, dyes, and medicines (increased);
paints, pigments, colors, and varnishes (increased) ; and paraffin
(decreased). (Pp. 919-21)
Statistics are given on the imports for consumption and
domestic exports of vegetable oil and vegetable-oil material
by British Dominions and Protectorates in Africa during the
three latest years for which statistics are available. Photostat
copies of detailed statistics showing countries of shipment of
imports and of destination of exports may be obtained from
the Bureau of Foreign and Domestic Commerce for 15 cents a
page. (Pp. 925-9)
The production of tar, rosin, and turpentine in Finland is de-
scribed. (P. 937)
The market for paraffin wax, stearic acid, and rosin in Argen-
tina is reviewed. (Pp. 940-1)
The discovery of new fire clay, copper, and salt mines is re-
ported in Azerbaijan. (P. 941)
Asphalt, which is reported to be very similar to the asphalt
deposits in Trinidad, has been discovered in Manitoba Province.
(P- 948)
Polish regulations relative to prices of crude oil and oil prod-
ucts are given. (P. 953)
The bauxite concessions in British Guiana have commenced
to produce a considerable supply of this mineral. (P. 956)
Special Supplements Issued in November
Finland — 6<j Spain — 1 8c
Portugal — 146 China — 55<f
Caucasus — 166 Japan— 58c
Canada — 266
Statistics op Exports to the United States
Belgium — (Pp. 518,
590)
Hides and skins
Wax
Copper
Minerals (unclassified)
Rubber
Resinous products
Chemicals
Bahia— (P. 583)
Hides and skins
Chrome ore
Manganese ore
Castor oil
Rubber
Medicinal roots and
Brazil— (Pp. 815,852)
Crude rubber
Bauxite
Great Britain — (P.
808)
Salt (not table)
Hides (undressed)
Skins
Cement, calcareous
Iron and steel
Lead
1 sulfate
Bleaching powder
Leather
Rubber, crude
South Australia —
(P. 797)
Yacca gum
London — (P. 853)
Rubber
Leather
Tin
Drugs and chemicals.
Gums
Lead
Aluminium
Ferromanganese
Creosote oil
Copper
Linseed oil
Scrap metal
Rubber
Naples-
Copper
Sulfur oi
(P. 773)
BOOK RE.VILW5
Application of Dyestuffs. By J. Merritt Matthews, xvi + 768
pp. John Wiley & Sons, Inc., New York, 1920. Price, $10.00.
The introduction of this work shows that it represents a
development and expansion of an earlier textbook for students
into a work of instruction and reference for those directly con-
cerned with the use of dyestuffs. In order to understand the
scope of this work it should be stated that it is definitely not
a book about the manufacture, constitution, or chemical classi-
fication of dyestuffs. These things are dealt with only as they
immediately concern the subject matter.
The book deals first with the effect of acids, alkalies, chemicals,
etc., on the textile fibers, and then with the methods of cleaning
and bleaching them, covering these matters fully, so far at least
as knowledge regarding them is likely to be valuable to the user
of dyes.
It then proceeds, after a relatively short and elementary
discussion of the classification of dyes, to its real subject, and
takes up fully their application by the usual methods to the
several textile fibers, and their construction. This occupies the
major part of the book, and is succeeded by a chapter on the
theory of dyeing, containing a most interesting presentation
and discussion of the current views on this subject. Then follows
consideration of fastness tests and chapters devoted to special
materials, not textile, such as straw, leather, paper, etc., and to
lakes and inks. The remainder of the book takes up testing of
dyestuffs, and their chemical reactions, the analysis of textile
fabrics, and such data and tables as are likely to be useful,
ending with a bibliography of value to those who wish to follow
up the literature of the subject. Along with the text are copious
footnotes, which carry a surprising amount of most valuable
information.
The development of this work from a textbook for students
has brought about the presence of a very desirable feature from
the point of view of the chemist of limited textile experience.
We refer to the general illustration of the important points by
very definitely described experiments. These are attractive
both to the beginner and to anyone who wants to have at hand
directions for demonstrating clearly in a laboratory what he
may already know.
It is difficult to criticize a book which is filled with such a deal
of information, but perhaps a few suggestions might be offered.
The writer of this review is particularly interested in the dyeing
of men's wear, and would have been glad to see a larger place
given to the merits and difficulties of the several classes of fast
chrome and mordant colors, and those auxiliary colors which
are used with them. For the man who has to deal with modern
schemes of piece dyeing, a treatment of resist work on woolen
or worsted yarn, and a discussion of silk dyeing in fast colors
for fulling and cross dyeing in men's wear would have been useful.
But perhaps Dr. Matthews felt that a limit must be placed
even in a work as broad as he has given us.
W. D. Livermore
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 1
The Chemists' Year Book, 1920. By F. W. Atack, M.Sc. Tech.
(Manch.), B.Sc. (London); Fellow of the Institute of Chem-
istry. Assisted by L. Winyates, A.M.C.T., A.I.C. 5th
Ed., 1 136 pp. in two volumes. Illustrated. Sherratt and
Hughes, London; Longmans, Green & Co., New York, 1920.
Price, $7.00 net.
These little volumes have been issued yearly since January
1915. They constitute another of the many illustrations of
British thoroughness and ability to finish the job that have been
given to us since the "contemptible little army" crossed the
channel a few months before the first issue of this Year Book
was forced by the cutting off of the supply of the well-known
German Chemiker Kalendar. Five years of success and im-
provement since then should constitute the work a permanent,
ready-reference landmark with its information in a handy
and easily accessible form.
This fifth edition, besides the usual general revisions and those
of the section on dairy products and carbohydrates, presents
its principal alteration in the complete recasting of the "Physical
Chemistry Constants" section by Dr. G. Barr of the National
Physical Laboratory.
The first volume is the smaller, 422 pages, and in general
embraces sections on atomic weights, with useful tables of
multiples, formula weights and their logarithms. Then comes
a very practical qualitative analysis section of 57 pages, in-
cluding treatment of some of the rarer elements. There are
sections on reagents, gravimetric analysis, volumetric, gas,
ultimate organic, electro- and spectrum analysis, invaluable
tables of general properties of inorganic and of organic sub-
stances, conversion tables of measurements, five-place loga-
rithms, and various mathematical constants.
The second volume of 714 pages embraces 180 pages of physical
constants, followed by an excellent illustrated section on crystal-
lography. The illustrations throughout are up-to-date and
helpful. Then follows a section on mineral properties, and a
long series of sections on technical analysis and control, including
water, fuel, efficiency of boiler plant, clays, cement, chemical
manufacture, oils, paint, agricultural chemistry, sugar, tanning,
textiles, dyes, intermediates, pharmaceuticals, trade names,
and constitution of synthetic drugs, rubber, and others. The
various special sections are written by specialists.
It is perhaps too much to expect from a "Chemists' Year
Book" many data on the engineering side of chemical production,
though the volumes are obviously intended for the industrial
chemist.
The usefulness of the many specific gravity-composition of solu-
tion tables is obvious. It is not so obvious, however, that our use
of them involves grave danger when unacquainted with the in-
dustrial status of the solution. The table of strength of formalde-
hyde solutions, for instance, would be very satisfactory if such
solutions did not always contain methanol as a preserva-
tive. Under the circumstances, the table is commercially
useless.
Some few things are a little hard to understand, such as the
fact that the sole reference to an original in the section on electro-
analysis is to a German publication, when the best work in the
field appears in our own journals as the work of Provost E. F.
Smith. That rotating electrodes will give more rapid results
is mentioned, but all data given are for stationary electrodes.
The use of warm hydrochloric acid to remove manganese dioxide
from platinum seems to demand care on the part of the nascent
chlorine liberated.
Citation of references to authority is not so frequent as might
have been. Omissions are sometimes glaring, as when a
brief table is cited from Colman for toluene evaluation (p. 957),
followed without any credit at all by two tables (pp. 959, 960),
which are precisely identical with those of F. E. Dodge in Rogers'
"Industrial Chemistry," with the exception of the typographical
error (p. 960) of 2 per cent at 1290 inste A of 1 per cent on the
20 to 80 "toluene-xylene" mixture. Ne Ve'theless, the work is
remarkably free from typographical error
The authors have done well in elimina jni: the needless diaiy-
calendar feature of the old Chemiker Kaiendar. The electro-
analysis section is more practical. In th'. ureful table of organic
compounds the use of the heading "formula weight" is to be
commended, but there is a little too much space taken up with
structural formulas, and the omission of the column of color,
crystal form, etc., to insert one of empirical formulas is a blunder.
Anyone can add up the empirical formula of a compound whose
structural formula is given, but not even an organic chemist
can imagine the crystal form and color of an unfamiliar com-
pound.
The work is not only well edited, but as a piece of book making
it is a model. The paper is good, and the print and make-up
are clean-cut and refreshing. James R. Withrow
The Microbiology and Microanalysis of Foods. By Albert
Schneider. 8vo x + 262 pp. 131 illustrations. P. Blak-
iston's Son & Co., Philadelphia, Pa. Price, $>3-50 net.
The author states in the preface: "This volume is intended
as a guide to the study of microbiological decomposition changes
in foods. It also presents a practical working basis for as-
certaining the decomposition limits of foods suitable for human
consumption, by means of the direct methods of microanalysis,
***** The text is addressed to army dietitians and food
examiners." Although the title of the book is given as "The
Microbiology and Microanalysis of Foods," the bulk of its
pages are devoted to what may be called food hygiene. Where,
however, the author treats of "microanalytic" methods, he
does so clearly and concisely, and all food analysts will welcome
this contribution to our knowledge of an intricate and puzzling
field which is sadly lacking in text and reference books.
If we are to accept the author's standards qualifying a man
to call himself properly trained to undertake investigations in
food and drug microscopy we must conclude that there prob-
ably does not exist a single individual in the United States who
can meet the requirements, for we are told that in addition to a
university training or its equivalent,
He must have made careful microscopical examinations of
all substances which may be so examined and that includes prac-
tically everything of a material nature. Skilled microanalysts
are rare. There are, indeed, many students who have been
taught certain things about the microscope and who have ex-
amined and reported upon certain microscopic objects and there
are many bacteriologists, biologists, chemists, and other in-
vestigators who make occasional use of the compound micro-
scope but these are not microanalysts in the true sense of the
term. The army microanalyst must be able to recognize at a
single glance all of the objects which may appear within any
field of the compound microscope.
It is no doubt in substantiation of the idea that microanalysts
must have studied "practically everything of a material nature"
that the author has introduced illustrations and diagrams which
are wholly irrelevant and to which no references are made in
the text, thus cutting down valuable space which might have
been used to good advantage in elaborating topics which had to
be discussed with but slight consideration.
The first six chapters or sections, comprising 68 pages, are
devoted to food hygiene, microbiology, and food decomposition,
and the statements of facts are as brief as it has been possible
to make them, and are, on the whole, correct. Reference to
authorities are unfortunately wholly omitted.
Chapter VIII (70 pages) is devoted to "General and Special
Microanalytical Methods." This chapter is a direct and valu-
able contribution to our literature of the microscopy of foods,
and will prove most acceptable to all microscopists who have
occasion to make microbiological examinations or who are re-
quired to undertake microscopic quantitative analyses. The
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
various methods which have been suggested for direct bacterial
counts in food and beverages are outlined, and the principles
underlying microscopic quantitative analyses are discussed at
length, together with the basis for the interpretation of the
results obtained.
The author devotes some 64 pages (Chapter IX) to the dis-
cussion of the interpretation of the results obtained by micro-
scopic examinations and states his views relative to the "Mi-
croanalytical Rating of Food Products." With many of these
ratings few analysts will agree, tb is being especially true of both the
methods for the examination and the ratingsof water and of gelatin.
It is to be regretted that the author in giving his ratings does
not state that under certain conditions the ratings given are
rather ideal and may prove impracticable of enforcement.
As a further guide to assist the analyst in passing upon the purity
of food, a compilation has been made in Chapter X of the legal
standards of purity of foods.
The typography and the general arrangement of the book are
excellent. The cuts are for the most part clear, and those which
have a direct bearing upon the subject matter of the text are
well chosen. E. M. Chamot
Fuel Oil in Industry. By Stephen O. Andros. 274 pp. The
Shaw Publishing Co., 910 So. Michigan Boulevard, Chicago,
111., 1920. Price, $3.75.
This is a comprehensive treatise, embracing the storage of
fuel oil, heating, straining, pumping, regulating, boiler furnace
arrangement, types of fuel-oil burners, fuel oil in steam naviga-
tion, oil-burning locomotives, use of oil in the iron and steel
industries, in heat treating furnaces, in the production of elec-
tricity, in the sugar, glass, and ceramic industries, the heating
of public buildings, hotels, and residences, and the use of oil in
gas making. In view of the enormous size of the fuel-oil industry,
there is no question but what there is place for a treatise on fuel
oil such as this book presents.
Because of the threatened shortage of petroleum, it is re-
grettable that so much of our petroleum is turned into fuel oil
instead of the more valuable products — gasoline, kerosene,
lubricating oils, wax, etc., but when, as the author says, with
equivalent bunker space, the use of oil over coal increases the
radius of action of ships over 80 per cent, and the M. K. and
T. R. R. in 1920 saved one-fourth of its fuel bill by using oil
instead of coal, the national and commercial reasons for using
oil as fuel are understood.
A chapter is devoted to colloidal fuel ; in the author's definition,
a combination of liquid hydrocarbons with pulverized carbona-
ceous substances (coal), the components so combined and so
treated as to form a stable fuel capable of being atomized and
burned in a furnace. As the author states, the title is not
scientific, since much of the solid component is not reduced
to colloidal dimensions. A reader naturally looks in the book
for a critical survey of the commercial status of colloidal fuel,
but does not find it, presumably because the substance has
scarcely passed the experimental stage. Eight pages are de-
voted to a description of the substance and to tests conducted
largely by Messrs. Dow and Smith, chemical engineers of New
York City. They made the interesting observation that in
some of the material 2.6 per cent of the particles became de-
stabilized (settled out) in 5 mo.' time. The author states
that 40 per cent by weight of coal can be suspended with 60
per cent by weight of oil, that the coal should be reduced so
95 per cent passes through a 100-mesh screen and 85 per cent
through a 200-mesh screen, and that the calorific value of the
fuel may be greater per unit volume than that of straight oil,
in some cases 15 per cent greater.
From a chemist's standpoint, the first chapter on principles
of fuel-oil combustion is not couched in language always scien-
tific, although clear and readable and perhaps well understood
by engineers who are not chemists. For instance, the author
states that copper wire is placed in cuprous chloride Orsat
pipets to reenergize the solutions if they become weakened.
The second chapter is devoted to properties and chemical
and physical tests of fuel oil. The tests are well selected and
described.
In the third chapter is found a comprehensive comparison
of fuel oil and coal. Analyses of coals are shown, also combus-
tion tests, costs of pulverizing coal, comparative efficiencies,
all well selected data, and finally a page and a half on advan-
tages and disadvantages of liquid fuel. One can find no fault
with this comparison.
A chapter on distribution and storage of fuel oil covers the
storage of fuel in ships, in locomotives, and on land, and above
and below ground. Concrete and steel construction are dis-
cussed. Regulations of the National Fire Protective Association
and the cities of New York and Chicago are included. A rule
of the New York City regulations provides that the fuel oil
must not be over 20° Be. This shuts out Mexican fuel oil.
The reviewer protested against this when the regulations went
into effect, but he could not discover the particular motive
behind it. However, there has been such an urgent demand from
other sources for Mexican fuel oil that apparently the producer
does not care.
The succeeding chapters, including one on heating, straining,
pumping, and regulating, are devoted tc appliances such as
boilers, burners, and locomotives, and to the use of fuel oil in
the various industries.
The chapter on the use of gas oil in gas making was probably
written before the gas makers of the country were thrown into
a near panic because of the recent big advance in gas oil prices,
else the author might have included some cost data.
There is no question that the book is a good treatise on the
subject and fills a much-needed want on up-to-date practice.
It should be in demand. George A. Burreu,
Analysis of Paint Vehicles, Japans, and Varnishes. By Clif-
ford Dyer HollEY. ix -f- 203 pp. John Wiley & Sons,
Inc., New York; Chapman and Hall, London, England, 1920.
Price, $2.50 postpaid, or 13s. 6d. net.
Professor Holley has written a singularly useful and needed
book, dealing with volatile thinners, paint oils, dryers, water in
paints, and the effect of storage, and containing chapters on
baking japans and varnishes which are remarkable, in the litera-
ture of the subject, for common-sense and practical value.
It is a compendium of the standard methods of analysis, where
there are any, and, lacking these, of what appears to the author
the best, though perhaps imperfect, methods available; written
with clearness and sufficient detail, and generally accompanied
with intelligible discussions of the problems involved. Prac-
tical experience in making paints is the only foundation for a
reasonable and just valuation of the various questions, and the
analyst who has this book on his desk will find many of his
troubles simplified, while the factory superintendent who is also
a chemist — the number is increasing— will get help in under-
standing what he is doing.
The book is particularly valuable for its numerous tables,
most of which are not new, but from widely scattered sources;
and, while there are plenty of references to original papers,
it is not needful to look them up, because their methods are
given in full, except in the case of some U. S. Government publica-
tions, which may be secured without cost, in cases where an
elaborate description is wanted.
One may not agree with Professor Holley about everything,
but there is no question of his sincerity and thoroughness, and
the book is most satisfactory. It is admirably printed, and
free, so far as this reviewer has discovered, from errors.
A. H. Sarin
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. i3> No. i
NLW PUBLICATIONS
Chemistry: La Chimie et la Guerre: Science et Avenir. Charles Moureu.
384 pp. Price, 10 fr. Masson & Cie., Paris.
Chemistry and Civilization. Allerton S. Cushman. 151 pp. Price,
$2.50. Richard G. Badger, Boston.
Colloids: Les Colloides. J. Duclaux. 288 pp. Gauthier-Villars & Cie.,
Paris.
Dictionary of Chemical Terms. James F. Couch. 214 pp. Price, $2.50.
D. Van Nostrand Co., New York.
Eminent Chemists of Our Time. Benjamin Harrow. 248 pp. Price,
$2.50. D. Van Nostrand Co., New York.
Handbook of Industrial Oil Engineering. John Rome Battle. 1131 pp.
h Illustrated. J. B. Lippincott Co., Philadelphia.
Logarithmic and Trigonometric Tables. Earle Raymond Hedrick. Re-
vised edition. 143 pp. Price, $1.40. The Macmillan Co., New York.
Lubricants: American Lubricants from the Standpoint of the Consumer.
L. B. Lockhart. 2nd edition, revised and enlarged. 341 pp. Price,
$4.00. The Chemical Publishing Co., Easton, Pa.
Oils: The Manufacture, Refining and Analysis of Animal and Vegetable
Oils, Fats and Waxes, Including the Manufacture of Candles, Margarine,
and Butter. Geoffrey Martin. 200 pp. Price, I2s. 6d. net. Crosby,
Lockwood & Son, London.
Organic Chemistry: Laboratory Experiments in Organic Chemistry. E. P.
Cook. 83 pp. Price, $1.00. P. Blakiston's Son & Co., Philadelphia.
Organic Chemistry: Theoretical Organic Chemistry. Julius B. Cohen.
New Ed. 604 pp. Price, $2.50. The Macmillan Co., New York.
Paint: Papers on Paint and Varnish and the Materials Used in Their Manu-
facture. Henry A. Gardner. 501 pp. Illustrated. Price, $10.00
net. P. H. Butler, 1845 B St., N. W., Washington, D. C.
Rubber: Plantation Rubber and the Testing of Rubber. G. Stafford
Whitby. (Monographs on Industrial Chemistry.) 559 pp. Price,
$9.50. Longmans, Green & Co., New York.
NEW JOURNAL
The Glass Industry. Published monthly by Glas
Co., Inc., 19 Liberty St., New York City. Price,
Industry Publishing
2.00 a year.
RECENT JOURNAL ARTICLES
Alcohol Distillation from Molasses. George M. Appell. Sugar, Vol. 22
(1920), No. 12, pp. 703-705.
Autoclaves: The Use of Autoclaves in Chemical Processes. Thomas
Callan. The Chemical Age (London), Vol. 3 (1920), No. 73, pp. 502-
503.
Blast Furnace: Heat Supplied to the Blast Furnace. W. W. Hollings.
The Blast Furnace and Steel Plant, Vol. 8 (1920), No. 12, pp. 651-655.
Brass: Investigation of Brass Foundry Flux. C. W. Hnx, T. B. Thomas
and W. B. Vietz. The Metal Industry, Vol. 18 (1920), No. 11, pp. 498-
502.
Chemical Standard Samples. C. O. Bannister. Journal of the Society
of Chemical Industry, Vol. 39 (1920), No. 20, pp. 351r-352r.
Chemicals: Safeguarding the Purchase of Chemicals. F. W. RussE.
Textile World Journal, Vol. 58 (1920), No. 24, pp. 41-43.
Coal: How to Test the Fitness of Coals for Coking. G. H. Elmore. Coal
Age, Vol. 18 (1920), No. 23, pp. 1130-1134.
Colloidal Fuel: The Case for Colloidal Fuel. Lindon W. Bates. The
Chemical Age (London), Vol. 3 (1920), No. 75, pp. 558-559.
Combustion: Killing Combustion Losses. Robert June. Combustion,
Vol. 3 (1920), No. 6, pp. 32-36.
Copper Losses in Slags. Frank E. Lathe. Engineering and Mining
Journal, Vol. HO (1920), No. 23, pp. 1076-1080.
Cost Accounting in the Rubber Industry. Ferd G. Kirby. India Rubber
World, Vol. 63 (1920), No. 3, pp. 155-157.
Cost Accounting: Relationship of Cost Accounting to Business Management
in Chemical Manufacture. William B. Ferguson. Chemical Age,
Vol. 28 (1920), No. 11, pp. 413-415.
Dyes and the Varnish Industry. George H. Priest. American Dyestuff
Reporter, Vol. 7 (1920), No. 23, pp. 19, 20.
Dyes: Tartar Emetic in the Dye House. Elton R. Darling. Textile
World Journal, Vol. 58 (1920), No. 22 pp. 47-49.
Dyestuff Industry in Relation to National Economics. L. W. Alwyn-
Schmidt. Color Trade Journal, Vol. 7 (1920), No. 6, pp. 167-170.
Electrolytic Zinc Methods. Herbert R. Hanley. Mining and Scientific
Press, Vol. 121 (1920), No. 23, pp. 795-804.
Explosives: Commercial Explosives and Their Uses. A. J. Strane. The
Coal Industry, Vol. 3 (1920), No. 11, pp. 526-528.
Fertilizers: The Determination of Borax in Fertilizers and Fertilizer Ma-
terials. William H. Ross. The American Fertilizer, Vol. 53 (1920),
No. 12, pp. 64-66.
Fertilizers: The Distillation Method for Estimating Borax in Mixed Fer-
tilizers. J. M. Bartlett. The American Fertilizer, Vol. 53 (1920),
No. 12, pp. 68-71.
Flotation: A Resume of Literature on the Theory of Flotation, with Critical
Notes. H. R. Adam. Mining and Scientific Press, Vol. 121 (1920),
No. 22, pp. 765-770. Reprinted from the Journal of the Chemical,
Metallurgical and Mining Society of South Africa.
Furnaces: Electric vs. Combustion Furnaces for Low Temperatures.
Frank W. Brooke and George P. Mills. Chemical and Metallurgical
Engineering, Vol. 23 (1920), No. 21, pp. 1008-1010.
Glass: The Annealing of Glass. L. H. Adams and E. D. Williamson.
Journal of the Franklin Institute, Vol. 190 (1920), No. 5, pp. 597-631.
Glass: The Composition of Lead Glasses. Robert J. Montgomery
Journal of the American Ceramic Society, Vol. 3 (1920), No. 11, pp. 900-912.
Glass: The Importance of Technical Control in the Glass Industry. Alex-
ander Silverman. The Glass Industry, Vol. 1 (1920), No. 1, pp. 1-3.
Glass: Some Developments in the Study of Glass Technology in the Year
1919-1920. W. E. S. Turner. Journal of the Society of Glass Tech-
nology, Vol. 4 (1920), No 15, pp. 287-296.
Glass: Some Principles in Glass Pot Manufacture. C. H. Kerr. The
Glass Industry, Vol. 1 (1920), No. 1, pp 3-6.
Leather: Hide Substance. G. L. Terrasse. Journal of the American
Leather Chemists' Association, Vol. 15 (1920), No. 11, pp. 608-620.
Leather: Rapid Estimation of Chromium in Chrome Liquor by Use of the
Immersion Refractometer, Arthur W. Thomas and Margaret W.
Kelly. Journal of the American Leather Chemists' Association, Vol. 15
(1920), No. 12, pp. 665-668.
Lime Treatment for Soil Acidity. A. T. Wiancko. Sugar, Vol. 22 (1920),
No. 11, pp. 650-651.
Mathematics: Le Role des Math6matiques en Chimie. E. Flburent.
Chimie et Industrie, Vol. 4 (1920), No. 4, pp. 525-528.
Medicinals: Les Medicaments Allemands. Augusts Lumiere. Chimie
et Industrie, Vol. 4 (1920), No. 4, pp. 435-442.
Metallurgy from the Colloid Chemical Standpoint. Clifford W. Nash.
Chemical Engineering and Mining Review, Vol. 13 (1920), No. 1 45, pp. 1 1-13.
Oil: A World View of the Oil Supply. George Otis Smith. Engineering
and Mining Journal, Vol. 110 (1920), No. 22, pp. 1047-1049.
Oils: L'Hydrogenation des Huiles. M. Frejacqub. Chimie et Industrie,
Vol. 4 (1920), No. 4, pp. 443-456.
Optical Glass: Certain Relations between Chemical Composition and Re-
fractivity in Optical Glasses. Fred. E. Wright. Journal of the Ameri-
can Ceramic Society, Vol. 3 (1920), No. 10, pp. 783-832.
Paper: Utilization of Reed in Paper Manufacture. Enn. Heuser. Paper,
Vol. 17 (1920), No. 9, pp. 23, 32.
Petroleum: Refining Problems of the Petroleum Industry. H. H. Hill.
Chemical Age, Vol. 28 (1920), No. 11, pp. 425-427.
Photography: The Measurement of Graininess in Photographic Deposits.
Loyd A. Jones and Noel Deisch. Journal of the Franklin Institute,
Vol. 190 (1920), No. 5. pp. 657-683.
Photo-Micrographs in the Textile Industry. W. F. Edwards. Textile
World Journal, Vol. 43 (1920), No. 20, pp. 43-47.
Rubber: Determination of Antimony in Rubber Goods. S. Collier, M.
Levin and J. A. Scherrer. The Rubber Age, Voi. 8 (1920), No. 3, pp.
104-105.
Rubber: Notes on Rubber Analysis. A. R. Pearson. The Analyst,
Vol. 45 (1920), No. 536, pp. 405-409.
Soap: The Surface Tension of Certain Soap Solutions and Their Emulsifying
Powers. Mollis G. White and J. W. Marden. Journal of Physical
Chemistry, Vol. 24 (1920), No. 8, pp. 617-629.
Steel: The Heat Treatment of Automobile Steels. Robert R. Abbot.
American Drop Forger, Vol. 6 (1920), No. 11, pp. 536-539.
Steel: Some Notes on the Effect of Nitrogen on Steel. O. A. Knight and-
H. B. Northrup. Chemical and Metallurgical Engineering, Vol. 23 (1920),
No. 23, pp 1107-1111.
Steel for Valves of Combustion Motors. G. Gabriel. The Iron Age, Vol.
106 (1920), No. 20, pp. 1249-1251; No. 23, pp. 1465-1469. Translated
from La Technique Automobile et Aerienne.
Steel: Study of the Testing of Welds. S. W. Mn.LER. American Drop-
Forger, Vol. 6 (1920), No 11, pp. 549-554.
Sugar: A New System of Cane juice Clarification. I. H. Morse. Louisi-
ana Planter and Sugar Manufacturer, Vol. 45 (1920), No. 19, pp. 301-302;
No. 20, pp. 315-317.
Sugar: On the Settling of Precipitates in General and of Cane Juice Pre-
cipitates in Particular. Noel Deerr. The International Sugar Journal,
Vol. 22 (1920), No. 623, pp. 618-624.
Textile Research: Modifying Influences in Textile Respaich. Louis A.
Olney. American Dyestuff Reporter, Vol. 7 (1920), No. 19, Section 2,
pp. 11-12.
Vitamines: A Quantitative Method for the Determination of Vitamine in.
Connection with Determinations of Vitamins in Glandular and Other
Tissues. Frederick K. Swoboda. Journal of Biological Chemistry,
Vol. 44 (1920), No. 2, pp. 531-551.
Zinc: Idrometallurgia dei Mineral! di Zinco. G. Aichino. Giornale di
Chimica Industriale ed Applicala, Vol 2 (1920), No. 10, pp. 566-572.
Jan., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
MARKET REPORT— DECEMBER, 1920
FIRST-HAND PRICES FOR GOODS IN ORIGINAL PACKAGES PREVAILING IN THE NEW YORK MARKET
INORGANIC CHEMICALS
Acid, Boric, cryst., bbls lb.
Hydrochloric, com'l, 22° lb.
Hydriodic oz.
Nitric, 42° lb.
Phosphoric, 50% tech lb.
Sulfuric. C. P lb.
Chamber, 66° ton
Oleum 20% ton
Alum, ammonia, lump lb.
Aluminium Sulfate (iron-free) lb.
Ammonium Carbonate, pwd lb.
Ammonium Chloride, gran lb.
Ammonia Water, carboys, 26°. . . .lb.
Arsenic, white lb.
Barium Chloride ton
Nitrate lb.
Barytes, white ton
Bleaching Powd., 35%, Works, 100 lbs.
Borax, cryst., bbls lb.
Bromine, tech lb.
Calcium Chloride, fused ton
Chalk, precipitated, light lb.
China Clay, imported ton
Copper Sulfate 100 lbs.
Feldspar ton
Fuller's Earth 100 lbs.
Iodine, resublimed lb.
Lead Acetate, white crystals lb.
Nitrate lb.
Red American 100 lbs.
White American 100 lbs.
Lime Acetate 100 lbt.
Lithium Carbonate lb.
Magnesium Carbonate. Tech lb.
Magnesite ton
Mercury flask American 75 lbs.
Phosphorus, yellow lb.
Plaster of Paris 100 lbs.
Potassium Bichromate lb.
Bromide, Cryst lb.
Carbonate, calc, 80-85% lb.
Chlorate, cryst lb.
Hydroxide, 88-92% lb.
Iodide, bulk lb.
Nitrate lb.
Permanganate, U. S. P lb.
Salt Cake, Bulk ton
Silver Nitrate oz.
Soapstone, in bags ton
Soda Ash, 58%, bags 100 lbs.
Caustic, 76% 100 lbs.
Sodium Acetate lb.
Bicarbonate 100 lbs.
Bichromate lb.
Chlorate lb.
Cyanide lb.
Fluoride, technical lb.
Hyposulfite, bbls 106 lbs.
Nitrate, 95% 100 lbs.
Silicate, 40° lb.
Sulfide lb.
Bisulfite, powdered lb.
Strontium Nitrate lb.
Sulfur, Sowers 100 lbs.
Crude long ton
Talc, American, white ton
Tin Bichloride lb.
Oxide lb.
Zinc Chloride, U. S. P lb.
Oxide, bbls lb.
.01'/,
.19
.07»A
.22
.07
20.00
23.00
• 04«/4
■ 04>/i
.16
.111/,
30.00
4.00
.08i/i
.53
33.50
.05
18.00
7.00
8.00
1.00
4.00
.16
.15
.12>/i
.lOVl
2.50
1.50
72.00
55.00
.35
1.50
.22
.30
.18
3.00
.12
.60
30.00
.51
12.00
1.90
3.80
4.00
2.90
.Oli/i
.08
.07
.15
4.00
20.00
20.00
.19Vi
.50
.40
ORGANIC CHEMICALS
Acetanilide lb.
Acid, Acetic, 28 p. c 100 lbs.
Glacial lb.
Acetylsalicylic lb.
Benzoic, U. S. P., ex -toluene., lb.
Carbolic, cryst., U. S. P., drs. . .lb.
50- to 110-lb tins lb.
Citric, crystals, bbls lb.
.15
.01'/.
.07«/«
20.00
23.00
.04«/«
• ll'/s
75.00
18
.00
6
.50
8
.00
1
.00
4
.00
.16
.15
.12'/.
lO'/i
2
.00
1
.50
.12
72
.00
50
.00
.55
30.00
.46
12.00
1.80
3.70
.08 'A
3.00
.10
.10
.24
.16
4.00
2.85
.Oli/i
.08
.07
4.00
20.00
20.00
. 19'/.
.50
.40
3.25
.10>/l
Acid {Concluded)
Oxalic, cryst., bbls lb.
Pyrogallic, resublimed lb.
Salicylic, bulk, U. S. P lb.
Tartaric, crystals, U. S. P lb.
Trichloroacetic, U. S. P lb.
Acetone, drums lb.
Alcohol, denatured, 190 proof. . . .gal.
Ethyl, 190 proof gal.
Wood, Pure gal.
Amyl Acetate gal.
Camphor, Jap. refined lb.
Carbon Bisulfide lb.
Tetrachloride lb.
Chloroform, U. S. P lb.
Creosote, U. S. P lb.
Cresol, U. S. P lb.
Dextrin, corn lb.
Imported Potato lb.
Ether. U. S. P., cone, 100 lbs lb.
Formaldehyde lb.
Glycerol, dynamite, drums lb.
Pyridine gal.
Starch, corn 100 lbs.
Potato, Jap lb.
Rice lb.
Sago lb.
Dec. 1
2.35
2.35
.35
.35
.48
.45
4.40
4.40
.16
.13>/i
.90
.80
5.50
5.25
2.30
2.30
4.00
3.75
1.10
.90
.041/4
2.75
3.18
.06 Vi
OILS, WAXES, ETC.
Beeswax, pure, white lb.
Black Mineral Oil, 29 gravity gal.
Castor Oil, No. 3 lb.
Ceresin, yellow lb.
Corn Oil, crude lb.
Cottonseed Oil, crude, f. o. b. mill. .lb.
Menhaden Oil, crude (southern), .gal.
Neat's-foot Oil, 20' gal.
Paraffin, 128-130 m. p., ref lb.
Paraffin Oil, high viscosity gal.
Rosin, "F" Grade, 280 lbs bbl.
Rosin Oil, first run gal.
Shellac. T. N lb.
Spermaceti, cake lb.
Sperm Oil, bleached winter, 38°. . .gal.
Stearic Acid, double-pressed lb.
Tallow Oil, acidless gal.
Tar Oil, distilled gal.
Turpentine, spirits of gal.
Aluminium, No. 1, ingots lb.
Antimony, ordinary 100 lbs.
Bismuth lb.
Copper, electrolytic lb.
Lake lb.
Lead, N. Y lb.
Nickel, electrolytic lb.
Platinum, refined, soft oz.
Quicksilver, flask Amer 75 lbs ea.
Silver oz.
Tin lb.
Tungsten Wolframite per unit
Zinc, N. Y 100 lbs.
5.50
2.72
.13V.
.14
.051/j
.45
85.00
55.00
.74
.33 V.
6.50
5.75
FERTILIZER MATERIALS
Ammonium Sulfate export. . . 100 lbs.
Blood, dried, f. o. b. N. Y unit
Bone, 3 and 50, ground, raw ton
Calcium Cyanamide, unit of Am-
monia
Fish Scrap, domestic, dried, f. o. b.
works unit
Phosphate Rock, f. o. b. mine:
Florida Pebble, 68% ton
Tennessee, 78-80% ton
Potassium Muriate, 80% unit
Pyrites, furnace size, imported. . . . unit
Tankage, high-grade, f. o. b.
Chicago unit
4.00
5.10
45.00
6.85
11.00
2.00
1.65
.10V2
5.25
2.72
. 13>/.
85.00
50.00
6.50
5.75
4
.00
5
.00
6
.85
II
.00
2
.00
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
COAL-TAR CHEMICALS
Crudes
Anthracene, 80-85% lb
Benzene, Pure gal
Ccesol, U. S. P lb
Cresylic Acid, 97-99% gal
Naphthalene, flake lb
Phenol, drums lb
Toluene, Pure gal
Xylene, 2 deg. dist. range gal
Intermediates
Acids:
Anthranilic lb.
B lb.
Benzoic lb.
Broenner's lb.
Cleve's lb.
Gamma lb.
H.
,1b.
Metanilic lb.
Monosulfonic P lb.
Napthionic. crude lb
Nevile & Winther's lb.
Phthalic lb.
Picric lb.
Sulfanilic lb.
Tobias lb.
Aminoazo benzene lb.
Aniline Oil lb
For Red lb.
Aniline Salt lb.
Anthraquinone lb.
Benzaldehyde, tech lb.
U. S. P.
lb.
Benzidine (Base) lb.
Benzidine Sulfate lb.
Diaminophenol lb.
Dianisidine lb.
' p-Dichlorobenzene lb.
Diethylaniline lb.
Dimethylaniline , .lb.
Dinitrobenzene lb.
Dinitrotoluene lb.
Diphenylamine lb.
GSalt lb.
Hydroquinol lb.
Metol (Rhodol) lb
Monochlorobenzene lb.
Monoethylaniline lb.
a-Naphthylamine lb.
6-Naphthylamine (Sublimed) lb.
f>-Naphthol, dist lb.
m-Nitroaniline lb.
0-NitroaniIine lb.
Nitrobenzene, crude lb.
Rectified (Oil Mirbane) lb.
P-Nitrophenol lb.
P^Nitrosodimethylaniline lb.
o-Nltrotoluene lb.
0-Nitrotoluene lb.
m-Phenylenediamine lb.
p-Phenylenediamine lb.
Phthalic Anhydride lb.
Primuline (Base) lb.
RSalt lb.
Resorcinol. tech lb.
U. S. P lb.
Schaeffer Salt lb.
Sodium Naphthionate lb.
Tuiocar b anilide lb.
Tolidine (Base) lb.
Toluidine, mixed lb.
o-Toluidine lb.
m-Toluylenediamine lb.
0-Toluidine lb.
Xylidine, crude lb.
2.20
2.25
.70
1.75
2.00
3.75
1.65
1.70
3.25
.85
1.75
2.00
6.75
2.25
.42
.4?
COAL-TAR COLORS
Acid Colon
Black lb. 1.00
Blue lb. 2.00
2.20
2.25
.70
1.7.5
2.00
3.75
1.60
1.70
3.25
2.25
1.25
2.50
2.50
.45
.45
1.00
1.(10
1.00
1.00
.80
.80
5.50
5.5J
8.00
8.00
1.90
6.75
2.90
2.90
.25
.25
1.50
1.50
1.30
1.30
2.30
2.30
2.00
2.00
2.75
2.50
.75
.75
1.10
1.10
.60
.60
1.75
1.75
.44
.44
.33
.33
1.50
1.50
1.75
1.75
1.00
2.00
Acid Colors (Concluded)
Fuchsin lb.
Orange III lb.
Red lb.
Violet 10B lb.
Alkali Blue, domestic lb.
Imported lb.
Azo Carmine lb.
Azo Yellow lb.
Ery throsin lb.
Indigotin, cone lb.
Paste lb.
Naphthol Green lb.
Ponceau lb.
Scarlet 2R lb.
Direct Colors
Black lb.
Blue 2B lb.
Brown R lb.
Fast Red lb.
Yellow lb.
Violet, cone lb.
Chrysophenine, domestic lb.
Congo Red, 4B Type lb.
Primuline, domestic lb.
Oil Colors
Black lb.
Blue lb.
Orange lb.
Red III lb.
Scarlet lb.
Yellow lb.
Ntgrosine Oil. soluble lb.
Sulfur Colors
Black lb.
Blue, domestic lb.
Brown lb .
Green lb.
Yellow lb.
Chrome Colors
Alizarin Blue, bright lb.
Alizarin Red, 20% Paste lb.
Alizarin Yellow G lb.
Chrome Black, domestic lb.
Imported lb.
Chrome Blue lb.
Chrome Green, domestic lb.
Chrome Red lb.
Gallocyanin lb.
Basic Colors
Auramine, O, domestic lb.
Auramine, OO lb.
Bismarck Brown R lb.
Bismarck Brown G lb.
Chrysoidine R lb.
Chrysoidine Y lb.
Green Crystals, Brilliant lb.
Indigo, 20 p. c. paste lb.
Fuchsin Crystals, domestic lb.
Imported lb.
Magenta Acid, domestic lb.
Malachite Green, crystals lb.
Methylene Blue, tech lb
Methyl Violet 3 B lb
Nigrosine, spts. sol lb.
Water sol., blue lb.
Jet lb.
Phosphine G., domestic lb.
Rhodamine B, extra cone lb.
Victoria Blue, base, domestic lb .
Victoria Green lb
Victoria Red lb.
Victoria Yellow lb.
STRY
Vol. 13, No
Dec. 1
Dec. 15
2.50
2.50
.60
.60
1.30
1.30
6.50
6.50
S.50
5.50
8.00
8.00
4.00
4.00
2.00
2.00
12.00
12.00
3.00
3.00
1.50
1.50
1.95
1.95
1.25
1.25
1.00
1.00
1.00
1 .00
.70
.70
1.65
1.65
3.50
3.50
2.00
2.00
2.20
2.20
2.25
2.25
.70
.7C>
1.65
1.65
1.40
1 .40
1.65
1 .65
1.75
1 .75
1.70
1.70
7.75
7.75
1 .10
1.10
1.00
1 .00
1.25
1.25
2.20
2.20
2.50
2.50
2.00
2.00
2.00
2.00
2 80
2.80
2.50
2.50
4.15
4.15
6.00
6.00
12.00
12.00
4.25
4.25
4.50
4.50
2.75
2.75
3.50
3.50
.85
.85
.70
.70
.90
.90
7.00
7.00
40.00
40.00
6.00
6.00
6.00
6.00
7.00
7.00
7.00
7.00
^dfiQ c/ournal oP
INDUSTRIAL
& ENGINEERING
CHEMISTRY
'Published Monthly by The American Chemical Society
Editor: CHAS. H. HERTY
Assistant Editor: Lois W. Woodford
Advisory Board: H. E. Barnard
Chas. L. Reese
Editorial Offices:
One Madison Avenue, Room 343
New York City
Telephone: Gramercy 0613-0614
J. W. Beckman A. D. Little A. V. H. Mory
Geo. D. Rosengarten T. B. Wagner
Cable Address: JIECHEM
Advertising Dbpartmbnt:
170 Metropolitan Tower
New York City
Telephone: Gramercy 3880
Volume 13
FEBRUARY 1, 1921
No. 2
CONTENTS
The Society's President for 1921.
100
Editorials:
Elementary Economics 107
The Road to Demoralization 108
Thoughts Translated into Deeds 10S
Sowing Good Seed 109
The Race Is Not Always to the Swift 109
Original Papers:
Measurement of Vapor Pressures of Certain Potas-
sium Compounds. Daniel D. Jackson and Jerome
J. Morgan 110
Rubber Energy. Win. B. Wiegand 1 IS
Reactions of Accelerators during Vulcanization.
II — A Theory of Accelerators Based on the Forma-
tion of Polysulfides during Vulcanization. Win-
field Scott and C. W. Bedford 125
The Action of Certain Organic Accelerators in the
Vulcanization of Rubber — III. G. D. Kratz,
A. H. Flower and B. J. Shapiro 128
Cellulose Mucilage. Jessie E. Minor 131
The Preparation and Technical Uses of Furfural.
K. P. Monroe 133
Further Studies on Phenolic Hexamethylenetetra-
mine Compounds. Mortimer Harvey and L. H.
Baekeland 135
Studies on Bast Fibers. II — Cellulose in Bast Fibers.
Yoshisuke Uyeda 141
Laboratory and Plant:
Gasoline from Natural Gas. V — Hydrometer for
Small Amounts of Gasoline. R. P. Anderson and
C. E. Hinckley 144
A Cold Test Apparatus for Oils. G. H. P. Licht-
hardt 145
Titration Bench. W. A. Van Winkle 140
Addresses and Contributed Articles:
Refining Raw Sugars without Bone-Black. C E-
Coates 147
Research Problems in Colloid Chemistry. W. D.
Bancroft 153
Pekin Medal Award:
Willis R. Whitney. A. D. Little 158
Presentation Address. Charles F. Chandler 160
The Biggest Things in Chemistry. Willis R. Whitney. 161
Scientific Societies:
Plans for the Spring Meeting; Centenary of the
Founding of the Sciences of Electromagnetism and
Electrodynamics; Dr. Henry A. Bumstead; Nichols
Medal Award; John Scott Medal Award; Rumford
Medal Presentation; President Smith Addresses
Joint Meeting; Calendar of Meetings 166
Notes and Correspondence:
History of the Preparation and Properties of Pure
Phthalic Anhydride; The Ignition of Fire Engine
Hose when in Use; Repairing Iron Leaching Vats;
Vapor Composition of Alcohol- Water Mixtures;
The British Dye Bill; European Relief Council 107
Washington Letter 169
Paris Letter 171
Industrial Notes ■ 172
Personal Notes 17.3
Government Publications 175
Book Reviews 1 79
New Publications 182
Market Report 183
Subscription to non-members, $7.50; single copy, 75 cents, to members, 60 cents. Foreign postage, 75 cents, Canada, Cuba and Mexico excepted
Entered as Second-class Matter December 19, 1908, at the Post Office at Easton, Pa., under the Act of March 3, 1879.
Acceptance for mailing at special rate of postage provided for in Section 1103, Act of October 3, 1917, authorized July 13, 1918.
Subscriptions and claims for lost copies should be referred to Charles L. Parsons, Secretary. 1709 G Street, N. W.. Washington, D. C.
106
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
THE SOCIETY'S PRESIDENT FOR 192
EDGAR FAHS SMITH
Forty-five years ago the American Chemical Society
was founded, and just a quarter of a century has passed
since Edgar Fahs Smith was its president. The So-
ciety gives expression to its appreciation of his labors
by choosing him once more for the highest office in its
gift, and in doing so it places in tried and worthy hands
the leadership of its fortunes.
Few remain now who can recall the struggles and
discouragements of those early years. So faint was
the breathing at
times that it seemed
almost as if the
patient was at his
last gasp. There
were chemists scat-
tered here and there
over the land, but
most of them were
kept too busy to give
time to investiga-
tion. The teacher
had little assistance
with his classes, and
the practical side of
building up our in-
fant industries was
all-absorbing. Be-
sides, the Society's
Journal had to enter
the field of publica-
tion with first one,
then two other jour-
nals. All honor,
then; to those who
had heart of hope
and, with vision of
the future, kept up
the struggle. In
these days of leader-
ship in many fields
of investigation it
is well to pause a
while and think of
the sturdy pioneers
who blazed the way
and made this prog-
ress possible.
Among these pioneers none stands higher than our
new president, and no one has such a host of friends
nor is so well-beloved. A kindlier soul has never
walked among us. Counselor and friend to all who
needed him. lover of the truth whether it lay hidden
in the nature around him or in his fellow man, with
deep, abiding faith in all that was fine and noble and
true, he has stood throughout the years four-square
to every wind that blew. His friendship has been an
inspiration and a blessing to many.
It might seem unnecessary to recount the contribu-
tions of Dr. Smith in the building up of our science
but, perhaps, there are some among our thousands
of members who do not realize how much his labors
have meant to all of us and how they have strengthened
chemistry in America and kept fresh the story of its
beginnings.
It is a somewhat striking coincidence that Dr. Smith
began his life work as a teacher of chemistry in the
University of Pennsylvania in 1876, the same year in
which our Society
was founded. Life-
long contemporaries
they have been in
the work. Starting
as an instructor, he
rose through the
various grades to
head of the depart-
ment of chemistry,
then vice provost,
and lastly provost
of the University
retaining through-
out his devotion to
his science and faith-
fully answering to
the limits of his
strength the calls
that were made upon
him. It is difficult
to measure such an
influence as he has
exerted. The story is
known to those who
had the good for-
tune to study under
him. They admire
him, they love him,
and happy are they
if they pattern after
him. In all these
years he has been
a wise and helpful
counselor in the af-
fairs of the Society,
and has done much to
Edgar Fabs Smith, President American Cbemicau Societv promote its interests.
As a teacher, he has been helpful in introducing new
methods and in providing excellent textbooks. At
first these were translations from the most widely ac-
cepted foreign authors — as witness his several editions
of Richter's "Organic Chemistry," and the "Electro-
chemistry" of Oettel. In this line he was one of the
first to have a well-equipped electrochemical laboratory
and to drill his students in this increasingly important
branch, issuing several valuable guides and textbooks
of his own. He devised new methods of analysis and
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
107
greatly aided in introducing this valuable adjunct to
the laboratory practice of the day. All of which was
fitting on the part of one who held the chair of Robert
Hare, who constructed the first American electric
furnace.
The long list of his investigations helps to fill the
pages of our Journal and need not be detailed here.
Suffice it to say that his interests and his work lie in
many fields. Chief among them are electrochemistry,
the complex inorganic acids, the rare earths, and the
revision of those constants, if constants they be, the
atomic weights. In this latter field he has covered
about one-fourth of the known elements, and his work
ranks high. This is a monumental work in itself.
His latest work on the atomic weights of boron and
fluorine is a fine example of how such work should be
done.
The many-sided interests of this man are shown by
the caretaking, accurate, and very valuable work which
he has done as a historian. His activities in this line
may have been aroused by the fact that he occupied
the chair which had been held by Benjamin Rush, the
first professor of chemistry in America, and lives in
the historic city of Philadelphia, where in 1792 was
"instituted" the first chemical society in the world,
antedating by a half century the London Chemical
Society, the first to be established in Europe. Also, he
is a member and for some years was president of the
American Philosophical Society, which was founded
by Benjamin Franklin.
Surrounded by such historic memories he has made
the past live over again in a series of books for which
those of us who do honor to the men who paved the
way for our feet cannot be too grateful. Hare per-
forms over again for us his surprising experiments
with the oxyhydrogen blowpipe which he invented,
and Woodhouse, Cooper, and others tell of their dis-
couragements and achievements. And now in the
account of Priestley in America, which he has just
published, we catch an insight into the character of
that great discoverer, his limitations offset by his sur-
prising vision, which some of us who have read much
about him had never gained before.
To such tried and approved leadership we intrust
the reputation and future of the Society.
Chapel Hiu., N. C. FRANCIS P. VENABLE
EDITORIALS
ELEMENTARY ECONOMICS
Some are arguing that duty-free importation of
scientific apparatus by educational institutions will
mean a great saving in dollars and cents. But to
discuss the economic aspect of this question it is
necessary to shake one's self loose from memories of
pre-war conditions and remember that to-day we are
dwelling in a very much changed world. Before the war
Germany, thanks to an abundance of cheap, highly
skilled labor, placed upon the market chemical wares
at prices with which American manufacturers could
not compete. To-day Germany is faced with
the obligation of paying off during the next twenty-
five or thirty years an enormous reparations debt.
To do this Germany will sell goods in compe-
tition at absurdly low figures in order to destroy
war-born industries in other lands, while charging
exorbitant prices wherever she has a monopoly.
There is abundant evidence of the correctness
of this statement. In Science, November 26, 1920,
page 511, Professor James Lewis Howe complains
that the file of a journal which had been offered
him less than a year before for 3,000 marks has now
risen in price to 25,000 marks (though the exchange
value of the mark had meanwhile depreciated only 50
percent). Monopoly: — exorbitant charge! But Pro-
fessor Howe explains the situation in this same com-
munication, for he quotes from a German firm's letter
to an American customer:
"A word about prices. I take it from your name and con-
nections that you are of German family and am therefore pre-
pared to make most liberal terms. As you doubtless know, it
has been generally agreed in commercial circles here that all
articles sold to uitlanders, and especially to Americans, shall
be priced considerably higher than the same thing sold to our
fellow-citizens, the idea being to in this way recuperate to some
extent from our late overwhelming losses and to make our recent
enemies aid us in paying our most outrageous and crushing war
debt.
"This policy has been adopted en bloc by our associated. . . .
since some time. But as a fellow German, I am prepared to
let you have these goods at the Berlin price, this of course being
in all confidence, my most dear sir."
No camouflage about that — as long as it is in the
family.
Now take the other side of the picture. England
developed during the war a chemical glassware indus-
try:— competition! The London Morning Post of No-
vember 24, 1920, quotes the following conditions of
the British market at that time:
(Price to
Retailer)
1 ,000-cc. separating funnel 4s. Od.
400-cc. flat bottom flask Os. 6.5d.
500-cc. graduated flask 0s. 5d.
15-cc. bulb pipet Is. 3.5d.
Potash bulb Is. 9d.
Aneroid barometer 7s. 6d.
Chemical thermometer for testing acids. ... Is. 2d.
Clinical thermometer Os. 8.5d.
British
(Cost to
Jan«facturt
17s. 7d.
Os. 11.5d.
6s. 6d.
3s. 9d.
is. 6d.
20s. Od.
3s. Od.
2s. 4d.
Destructive competition! Do you believe those
German prices will stand after the British industry is
destroyed, say, four or five years, with that great
reparation debt still having twenty or twenty-five
years to run? We would be the veriest financial
babes-in-the-woods if we deliberately shut our eyes to
such a situation.
As further evidence, if it be needed, we quote from
The Chemical Age (London), December 25, 1920, in
summarizing the report of the Subcommittee on
Chemical Glassware appointed by the Standing Com-
mittee on Trusts:
"The nature of the foreign competition they have to meet
may be gathered from the fact that, favoured by exchange rates
108
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
and other conditions, goods of the kind now being made in this
country are being supplied by Continental manufacturers at
prices less than the actual cost of manufacture here, whereas
for goods that are nol yet being manufactured here prices are being
charged by the Continental makers which mean to the consumer
approximately five times the pre-war price of such goods.'' — [Italics
ours. ]
The U. S. Tariff Commission gives a new slant to
the whole question. In its report on chemical glass-
ware just submitted to the Ways and Means Com-
mittee (Tariff Information Surveys, Scientific Instru-
ments and Apparatus, page 59), it says:
"The great durability of domestic glassware makes it the
cheapest in the final analysis. Institutions which sell at actual
cost will no doubt find it to their advantage to use this material
regardless of the price of foreign ware, because, although the
first cost is high, the replacement cost is low and smaller reserve
stocks can be carried. Those institutions, on the other hand,
which plan on obtaining a profit from the sale of glassware to
students will find it to their advantage to use the fragile foreign
material. In this case heavy breakage increases the turnover
and therefore the profit."
The Tariff Commission is not disposed to joke, nor
to make charges without facts on which to base them.
Foster the American industry, then see that it plays
the game fair!
THE ROAD TO DEMORALIZATION
Two German dye" chemists, Dr. Otto Runger and
Dr. Joseph Flachslander, were officially released from
Ellis Island and admitted into this country on Janu-
ary 5, 1921. This action followed a thorough investi-
gation by the authorities of the port of New York
based, according to press accounts, upon a protest
from Germany. We don't blame Germany for pro-
testing, but with this side of the matter we have no
concern. The herrschaflen proceeded immediately to
Wilmington, Delaware, to take positions in the re-
search laboratories of the du Pont Company. Ac-
cording to the newspapers, $25,000 each is the salary
of these newcomers. Rumor has it that the amount
is much larger. A high official of the Company in-
forms us that these reports are greatly exaggerated.
However, that matter is not important. But the
changed policy of this Company, hitherto always
considered 100 per cent American in every respect, is
important, and unfortunate from whatever angle viewed.
An economic battle for the possession of the Ameri-
can market is in progress between the American and
the German dye industry. In war information is
obtained as far as possible from captured opponents,
but renegades are not placed in positions of high com-
mand. Whatever tends to demoralization in the
American ranks is a matter of national concern,
and the gravest feature of this new policy is the
lowered morale of the du Pont research staff which
will result therefrom.
It is not difficult to imagine the feelings of American
chemists who must take direction from men who
a short while ago were busy in those plants whence
came high explosives and poison gases, the latter ac-
counting for a full third of our hospital casualties.
Temperamentally that research staff now becomes a
conglomeration of incompatibles, a hybrid mixture
which has in it the elements of failure. At the outset
of the building of the dye industry there were many
laboratories where such a mixture was found to be
thoroughly bad, and where the weeding-out process
was put into operation and the staffs Americanized
with consequent fine results.
It is easy to understand the feeling of discourage-
ment which must possess the officials of the du Pont,
as of every other American dye manufacturing com-
pany, over the failure of Congress to enact definite
and adequate protective legislation. However, the
pressure from consumers for a wider variety of dyes
has been materially lessened through the constant
licensing of imports by the War Trade Board and by
the decreased demand for dyes during the present
general industrial slump. Now is the time for de-
veloping an efficient research staff from among our
ablest American chemists.
It is not too late to repair the damage. There are
eastward-bound steamers constantly traveling across the
Atlantic. Whatever the ability of these two chem-
ists, however intimate their knowledge of special lines
of manufacture may be — send them home and let the
American industry proceed to its full development
in an American way and by the force of American
brains.
THOUGHTS TRANSLATED INTO DEEDS
Often we discuss, and plan, and build great air
castles, and develop momentary boundless enthusiasm
— and then, with the peak of the curve reached, enthusi-
asm wanes, interest subsides or becomes diverted to
other matters, and the result is nothing. Happily for
progress this is not always the case.
At the meetings of the Interallied Conference of
Pure and Applied Chemistry which met in London
and Brussels, in July 1919, it was determined seriously
and comprehensively to set about the task of better-
ment of chemical literature. The American Chemical
Society undertook for its share of this work the prep-
aration and publication of two series of monographs,
scientific and technologic, on chemical subjects. The
announcement of the issuance of the first of the scien-
tific series "The Chemistry of Enzyme Actions" by
Dr. K. George Falk is an earnest that the American"
Chemical Society proposes to carry out promptly
and to the full its part of this undertaking.
Congratulations to the three trustees. Drs. Charles
L. Parsons, John E. Teeple, and Gellert Alleman, who
so quickly finished the business arrangements con-
nected with these publications; to the editors, Drs.
W. A. Noyes and John Johnston, who already have
announced progress in the preparation or printing of
eleven other monographs; and to the Chemical Catalog
Company, Inc., which has so excellently carried out the
publication of this first of the series.
Clear a new space on your book shelves, there is a
lot of fine material on the way to you!
Feb., 1921 THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
109
SOWING GOOD SEED
There have been strange doings in Washington. In
spite of the sentiment in Congress that the Chemical
Warfare Service should be developed to the fullest
extent, orders issued by high officials of the War De-
partment have tended to restrict its activities, to cripple
development, to prevent the training of troops in the
methods of gas warfare, in short, to limit the Chemical
Warfare Service solely to research.
Fortunately we are building for the future on better
lines, and in this work the American Chemical So-
ciety is doing a fine part through the annual lectures
given by distinguished members of the Society at the
United States Military and Naval Academies. The
first set of those lectures was given last winter, and
it will interest all to learn that of the graduating class
this year at West Point, 25 members requested as-
signment to the Chemical Warfare Service. The
second series of lectures is now in progress.
Recently we asked for frank opinions of the value
of these lectures. The Superintendent of the Military
Academy, Brigadier General MacArthur, wrote in
reply:
Through the courteous cooperation of the American Chemical
Society, following suggestions advanced in an editorial in the
Journal of Industrial and Engineering Chemistry for March 1919.
there were given last winter to the senior class of the Corps of
Cadets of the U. S. Military Academy a series of lectures on
important chemical processes. The lecturers and their subjects
were :
Dr. W. H. Nichols, "Sulfuric Acid, the Pig Iron of Chemistry"
Dr. C. L. Parsons, "The Fixation of Atmospheric Nitrogen"
Dr. W. H. Walker, "The Manufacture of Toxic Gases"
Dr. C. L. Reese, "Smokeless Powders and High Explosives"
Other lectures were planned but had to be omitted owing to
reduction in time made necessary by the war-time schedule
then being followed. These gentlemen, whose services were
entirely voluntary, placed their subjects before the class in an
extremely vivid, lucid and interesting manner, giving that
personal touch not to be found in textbooks and arousing the
keenest interest in their auditors, both by the subject matter
and by the manner in which it was presented.
The obvious benefit of these lectures has led to a continuation
of the policy and in the coming spring a second series will be
delivered, the lecturers and their proposed subjects being:
Dr. John Johnston, of Yale, "Industrial Research," March 23, 1921
Professor William McPherson, of Ohio State University, "Large
Scale Production of Munitions," March 30, 1921
Dr. G. A. Richter, of Berlin, N. H., "Rockets and Flares," April 6,
1921
Dr. G. W. Gray, of New York, N. Y., "Fuel. Motor and Lubricating
Oils," April 13. 1921
Dr. W. Lee Lewis, of Northwestern University, "Toxic Gases," April
20, 1921
Rear Admiral Scales, Superintendent of the Naval
Academy, was equally enthusiastic in his reply:
The suggestion for a series of lectures to be given at the
Naval Academy by members of the American Chemical So-
ciety first received public attention in an editorial entitled
"The Soldier, the Sailor and the Chemist" which appeared in
the Journal of Industrial and Engineering Chemistry for March
1919. The attention directed to this very important matter
aroused the interest of all concerned. The cordial offer of the
American Chemical Society, tendered by the President, Dr.
William H. Nichols, to arrange for a series of lectures was much
appreciated and the opportunity gladly made use of.
During the academic year 1919-20, eight lectures in the general
field of chemical engineering were delivered at Annapolis by
members of the American Chemical Society. All of these
lectures were heard by student officers attending the Naval
Postgraduate School and four of them by the First (senior)
Class of midshipmen. During the academic year 1920-21 a
series of six lectures has been arranged, all of them to be heard
by the student officers of the Postgraduate School and four of
them by the First Class of midshipmen. The lecturers for the
current session are:
Dr. John Johnston, "Industrial Research," December 4, 1920
Dr. A. S. Cushman, "Preservation of Iron and Steel," January 8, 1921
Dr. G. W. Gray, "Fuel, Motor and Lubricating Oils," Februarv 4
and 5, 1921
Dr. Wilder D. Bancroft, "Organized Research," March 4 and 5, 1921
Dr. W. Lee Lewis, "Toxic Gases," April 1 and 2, 1921
Dr. Charles L. Reese, "Explosives," April 29 and 30, 1921
The series of lectures of last year, and the current series, are
proving both interesting and profitable to all who have the op-
portunity of hearing them, as they gain at least a perspective
of what the profession of chemical engineering has done, and
can do, in furnishing indispensable assistance to our military
and naval forces in preparation for, and in conduct of, active
operations calculated to carry into effect the requirements of
our national views and aims.
It is clear to us that the purpose contained in the original
editorial suggestion is being accomplished. The ultimate
benefits of the cordial cooperation of the American Chemical
Society cannot be given a definite value, but it is certain that
the movement now under way cannot fail to be productive
of much good to the naval service.
Surely no more patriotic and fruitful work than
the delivery of these lectures could be done by the
members of the Society.
THE RACE IS NOT ALWAYS TO THE SWIFT
We hustling Americans are apt sometimes to poke
good-natured fun at the slowness of the Britisher.
But sometimes the shoe is on the other foot, witness
the following chronological history of the British
ten-year dye license bill in Parliament:
December 2, 1920
December 3, 1920
December 7, 1920
December 7, 1920
December 8-15, 1920
December 17, 1920
December 17, 1920
(midnight)
December 21, 1920
December 22, 1920
December 23. 1920
December 23, 1920
(midnight)
January 15, 1921
Bill introduced in House of Co
reading, ordered printed.
Bill printed, distributed and received endorse-
ment of Colour Users Association
London Times in a leading editorial said:
"Attack is threatened from irreconcilable Free
Traders [our Senator Thomas], out-and-out Pro-
tectionists [modified to straight-tarifl-proteetionists.
our Senator Moses], and a section of the textile
trade [our Mr. John P. Wood and his adherents]."
Continuing, the Times said in comparing with
other key industries: "There is justification for
giving the dye industry preference on the ground
that it is essential both from the economic and the
military standpoints."
Bill moved to second reading. While a member
was speaking in opposition, at eleven o'clock the
closure was moved and carried by 280 votes to 74.
The second reading was agreed to.
Bill considered in Committee.
Third reading of the bill and passage by 118
votes to 25.
First reading in the House of Lords.
Second reading, passed 83 to 36.
Passed Committee consideration.
Bill passed third reading in the House of Lords.
Bill received the royat assent.
Law became effective.
Nearly two years have elapsed since the Longworth
bill was introduced in Congress. It is still there.
What's the matter with us, anyhow?
Our correspondence basket is overflowing with a fine
crop of "Tell-it-to-Herty" communications. Indi-
vidual acknowledgment will eventually be made, mean-
while things are moving.
110
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
ORIGINAL PAPERS
NOTICE TO AUTHORS: All drawings should be made with
India ink, preferably on tracing cloth. If coordinate paper is
used, blue must be chosen, as all other colors blur on re-
duction. The larger squares, curves, etc., which will show in
the finished cut, are to be inked in.
Blue prints and photostats are not suitable for reproduction.
Lettering should be even, and large enough to reproduce
well when the drawing is reduced to the width of a single column
of This Journal, or less frequently to double column width.
Authors are requested to follow the Society's spellings on
drawings, e. g., sulfur, per cent, gage, etc.
MEASUREMENT OF VAPOR PRESSURES OF CERTAIN
POTASSIUM COMPOUNDS1
By Daniel D. Jackson and Jerome J. Morgan
Columbia University, New York, N. Y.
Received December 9, 1920
Anderson and Nestell,1*^ a report on "The Volatiliza-
tion of Potash from Cement Materials," give the pre-
dominating factors affecting the recovery of potash in
the furnace gases beyond the furnace, as follows:
(1) The temperature prevailing in the kiln; (2) volume of
gas passing; (3) the intimacy of contact between the furnace
gases and the cement mix; (4) the vapor pressure of the potash
salt or salts formed; (5) the possibility of dissociation under
certain furnace conditions (oxidizing, neutral, or reducing atmos-
phere or changing temperature) to components of greater or
less volatility than the original salt; (6) the degree of saturation
of the gas in contact with the cement material; (7) the rate of
diffusion both of the salt vaporizing in the interstices of the
cement mix to the surface of contact with the gas stream, and
of the saturated gas at the surface to the leaner gas areas beyond.
Of these seven factors, all may be more or less va-
ried at will except the fourth, namely, the vapor pressure
of the potash salt or salts formed. It was decided, there-
fore, that the fundamental thing in a study of the
volatilization of potash is the determination of the
vapor pressure of the potassium compounds involved.
In the present work results of vapor pressure measure-
ments are given for three natural silicates, leucite,
orthoclase feldspar, and glauconite, which are suffi-
ciently abundant to serve as sources of potash, and for
four other potassium compounds, the chloride, car-
bonate, hydroxide, and sulfate, which are of particular
interest on account of their connection with the recovery
of potash from cement mill flue dust. The knowledge
acquired in these vapor pressure measurements will
later be applied to the study of the volatilization of
potash from mixtures of silicates with releasing and
volatilizing agents.
PREVIOUS WORK
In I860, Bunsen2 determined the relative volatility
of certain salts by heating a centigram bead of the
salt on a platinum wire in the hottest part of a Bunsen
flame and measuring the time required for the salt to
volatilize. In 1897, Norton and Roth3 repeated and
1 Part of a thesis presented in partial fulfilment of the requirement for
the degree of Doctor of Philosophy in the Faculty of Pure Science, Columbia
University, New York, N. Y.
* Numbers refer to references at end of paper.
extended the work of Bunsen. The volatility of sodium
chloride thus measured in each case was taken as
unity. The results of these investigators, as far as
they relate to potassium compounds, are given in
Table I.
Table I — Volatility of Potassium Compounds, Taking the Volatility
of Sodium Chloride as Unity
Results of Results of
Compound Bunsen Norton and Roth
Iodide 2.828 2.362
Bromide 2.055 1.860
Chloride 1.288 1.083
Fluoride 0.329
Carbonate 0.310 0.277
Sulfate 0.127 0.149
Bergstrom,4 in 1915, found the boiling points of the
potassium halides to be as follows: potassium chloride
1500°, potassium bromide 1435°, and potassium iodide
1420°. Niggli5 found that a mixture of potassium
carbonate and silica heated for 60 hrs. at 900° to 1000°
lost 15 mg. of K20. In addition, many of the recent
articles dealing with processes for recovering potash
from silicates contain statements as to the relative
volatility of certain potassium compounds, but, with
the exception of the work of Anderson and Nestell,1
it is believed that there has been no previous quanti-
tative study on the volatilization of potassium com-
pounds.
METHOD OF VAPOR PRESSURE DETERMINATION
On account of the difficulty of finding a gastight
material which would withstand the corrosive action
of potassium compounds at high temperatures, and of
measuring small pressures at these temperatures, it
seemed useless to attempt to employ a static method
for measuring the vapor pressure. Hence the dynamic
method of von Wartenberg6 was chosen.
In this method a measured volume of gas is passed
over a weighed quantity of the substance whose vapor
pressure is to be determined at the desired temper-
ature. The amount volatilized is found by the loss
of weight, and the partial pressure is calculated from
the relation:
Moles of substance X total pressure
Pressure of substance = :
Moles of gas + moles of substance
This partial pressure of the volatilized substance repre-
sents its vapor pressure only if the gas passed over
the heated substance is saturated with the vapor of
the substance at the given temperature, a condition
which is never realized experimentally. However, the
degree of saturation of the gas stream is inversely
proportional to its speed. Hence by determining these
partial pressures at three or more speeds of the gas
stream, and plotting the partial pressures against the
speeds, it is possible to obtain the slope of the line
which shows the relation between partial pressures of
the volatilized substance and speed of the gas stream.
If this line is extended to zero speed it gives the par-
tial pressure at saturation, which is the vapor pressure
of the volatilized substance.
The application of this method presupposes a knowl-
edge of the molecular weight in the gaseous state of
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
both the substance volatilized and the gas used, in
order that the number of moles of each may be cal-
culated. The number of moles of nitrogen, the gas
passed through the reaction chamber, was easily found
by weighing the water displaced by the nitrogen at
a given temperature and pressure. In the case of the
potassium compounds volatilized, the density in the
gaseous state has been determined for only one of the
compounds studied. Nernst7 has shown that the molec-
ular weight of potassium chloride at high temperatures
corresponds to the simple formula KC1. In calculat-
ing the vapor pressures of the other compounds it was
necessary to make certain assumptions regarding the
molecular weight of the compound volatilized. The
details of these assumptions are given under the dis-
cussion of the results for each compound. It can be
pointed out here, however, that should later work
show that the assumed molecular weight in any case
is wrong, it will simply necessitate recalculation of
the results and will not impair the usefulness of the
experimental data. Furthermore, a vapor pressure
here given, used in connection with the assumed molec-
ular weight, will give practically the same result in
calculation of the amount of potash necessary to sat-
urate a given volume of gas at a given temperature
and pressure as would a corrected molecular weight
used with the recalculated vapor pressure. Never-
theless, to avoid misunderstanding special attention
is called to the fact that, with the exception of the
value for potassium chloride, the vapor pressures
herein reported are based upon assumed molecular
weights.
VAPOR PRESSURE APPARATUS
A general sketch of the apparatus is given in Fig. 1.
It consisted of the gas container A, the purifying
train B, the vapor pressure tube C, which was heated
in an electric furnace, F, the absorbing train D, and
the gas measuring apparatus E.
GENERAL SKETCH
OF
VAPOR PRESSURE APPARATUS
The gas, nitrogen, which was to be passed through
the vapor pressure tube was contained over water in
a large bottle, A, which was connected by a syphon
with another bottle, A', containing a supply of water.
This second bottle was suspended from a screw ele-
vator so that the pressure of the gas in the apparatus
could be kept constant within one centimeter of water
pressure during the course of an experiment. A small
manometer, M, filled with water showed the pressure
in the apparatus.
After leaving the gas container and before entering
the vapor pressure tube the gas was freed from any car-
bon dioxide which might be present by passing through
the soda lime tube b', and dried by passing through
the calcium chloride tube b" , of the purifying train B.
After leaving the vapor pressure tube the gas passed
through the absorbing train D, which consisted of
three U-tubes filled as follows: d\ granular anhydrous
calcium chloride; d", soda lime in the first leg and bend
and calcium chloride in the second leg; d"' , calcium
chloride. The object of this purifying train was to
prevent moisture from diffusing back into the vapor
pressure tube and to absorb for weighing carbon di-
oxide set free by heating potassium carbonate in the
determination of its vapor pressure.
The speed at which the gas was passed through the
vapor pressure tube was regulated by the size of the
capillary in the tip g, through which water was allowed
to flow from the bottle E, and the volume of gas
passed through the vapor pressure tube was determined
by weighing the water displaced. By using a bottle
with large cross-section and extending the outlet tube
/, 2 liters of gas could be drawn into the measuring
apparatus with a loss of only about 3 in. in a total
head of 40 in. This is a change of 7.5 per cent, but
experiments with different sizes of capillary tips showed
an extreme variation of about 6 per cent in the speed
of the water flowing during the first minute and during
the last minute. The speed of the gas stream, there-
fore, varied during the course of an experiment not
more than 3 per cent from the mean speed. The tube
h, connected with the outlet tube, was open at the top
and allowed the pressure in the measuring apparatus
to be read upon the scale i. The rubber stopper of
the bottle E had four holes and carried, besides the
inlet tube shown in the figure, a tube by which water
could be introduced and two thermometers, one to
show the temperature of the gas and the other that of
the water. In order to give as small variation as pos-
sible in the speed of the gas stream, before beginning
an experiment a weighed quantity of water was run
out and the level of the water brought below the shoul-
der of the bottle. The temperature of the gas at the
beginning and end of the experiment was noted and
correction made whenever necessary for the change of
volume due to change of temperature.
A longitudinal section of the vapor pressure tube
C is shown in Fig. 2. The tube was made of "Im-
pervite" porcelain, 24 in. long and 1 in. bore, with
walls about three-sixteenths inch thick. It was glazed
on the outside and was found to be gastight at the
temperatures employed. Into this tube was cemented
with a grout of impervite body the fixed plug of im-
pervite which was perforated with a one-sixteenth inch
hole and had a recess for the Pt — Pt + Ir thermo-
couple as shown. The loosely fitting plug was also
of impervite body, unglazed, and had embedded in it
a piece of platinum wire by which it could be with-
drawn from the tube. The diameter of this plug was
about one-sixteenth inch less than the internal diam-
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
Fig. 2 — Longitudinal Section op Central Portion of Vapor
Pressure Tubb
eter of the tube. Gas flowing through the vapor
pressure tube was heated by passing through the
space between the loosely fitting plug and the walls
of the tube, and after passing over the substance con-
tained in the platinum boat and taking up its load of
vapor left the reaction chamber by the one-sixteenth
inch hole in the fixed plug. The entrance end of the
tube, which projected about 7 in. from the furnace,
was closed by a rubber stopper carrying a glass tube
through which the gas was introduced. The exit end
of the vapor pressure tube, which projected from the
furnace only about 1 in., was closed by a special
stopper molded of a mixture of portland cement and
asbestos. This was doubly perforated and carried an
exit tube for the gas and a double-bored porcelain pro-
tecting tube for the platinum-iridium thermocouple.
It was cemented into the tube by a mixture of sodium
silicate and barium sulfate, and the joints were made
gastight by coating with Bakelite varnish. At the
higher temperatures. the ends of the vapor pressure
tube were cooled by strips of wet filter paper so that
there was no decomposition of the rubber stopper or
of the Bakelite varnish.
The vapor pressure tube was heated in a molyb-
denum-wound electric furnace, details of which are
given in Fig. 3. The position of the tube in the fur-
nace was such that the reaction chamber was in the
central evenly heated portion of the furnace. Evidence
that the reaction chamber was evenly heated is given
by the fact that when the loosely fitting plug was with-
drawn it was only after a few seconds that the out-
lines of the platinum boat became visible.
The temperature of the furnace was regulated by
suitable resistances and was controlled by means of
a platinum-iridium thermocouple connected with a
Siemens and Halske millivoltmeter. The hot junc-
tion of the thermocouple was located in the recess in
the fixed plug as shown in Figs. 2 and 3. The cold
junction connections of the couple wires with the cop-
per leads of the millivoltmeter were made in mercury,
which was kept at a constant temperature by a water
bath. The temperatures in the reaction chamber cor-
responding to readings on the millivoltmeter were de-
termined at the beginning of each set of experiments
by a platinum-rhodium couple and a Leeds and North-
rup service potentiometer.
By substituting for the regular loosely fitting plug
a perforated plug of the same size, the hot junction of
the platinum-rhodium couple was supported over the
empty platinum boat in the position indicated in Fig.
2. Gas was then run through the vapor pressure tube
just as in a regular experiment. The cold junction
connections of the platinum-rhodium couple with the
leads of the service potentiometer were silver soldered
and kept at 0° C. in a vacuum bottle packed with ice.
The temperature was calculated from the electro-
motive force read on the potentiometer by Holman's
formula,
e = wT",
using the values m = 0.00275 and n = 1.18, which were
obtained for this particular thermocouple by calibra-
tion against the freezing points of zinc, antimony, and
copper, by Mr. Roland P. Soule in the physics depart-
ment of Columbia University. It is thought that
these temperatures are correct within ±10° C., and
the variation of the temperature during the course of
an experiment was always well within these limits.
PROCEDURE
When the temperature in the tube, as shown by the
platinum-iridium couple, had become constant at the
required point, and a constant pressure of about 2
cm. of water showed that there was no leak in the
system, the loosely fitting plug was withdrawn, a
platinum boat containing a weighed amount of potas-
sium salt was introduced, the plug quickly replaced,
and the gas stream through the tube started by allow-
ing water to run from the capillary tip g (Fig. 1)
into a weighed container. The temperature in the
tube was read at 3- to 5-min. intervals, and kept con-
stant within =*=5°; the pressure in the system was
kept constant within ±0.5 cm. of water by raising
the syphon bottle of the gas container. After about
2 liters of gas had been drawn through the tube the
gas stream was interrupted and the boat containing
the potassium salt quickly removed.
*__ „ „ __J
3^
I ,rurT»ce S»«l of '/e'Srrf
Fig. 3 — Section op Molybdenum- Wound Electric Furnace
A — Alundum Core, 10' X 2" Bore, Wound with 27 Ft. 0.028" Molyb-
denum Wire
Core, 12" X 5" Bore S — Electric Connector ol «/«"
Steel Rod
— No. 10, Copper Feed Wire
E — Alundu
K — Alundum Cement Rings
X — Rings of l/i" Asbestos Wood
P— Asbestos Fire Felt. '/<" Thick
R — Leads of Molybdenum Wrire,
4 Ply
U— Glass "T" Tube
V — Porcelain Insulating Tube
X— Rubber Tubing
The time between starting and stopping the gas
stream was noted, as well as the temperature of the
gas in the measuring apparatus and the pressure in
the apparatus. The volume of gas at this temperature
and pressure and saturated with water vapor was
found by weighing the water displaced, its volume
under standard conditions and dry was calculated,
and from this the number of moles of gas passed through
the vapor pressure tube was determined. The amount
of potassium compound volatilized was found either
by loss of weight or by analysis. All weighings were
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
113
corrected to actual grams mass, and the total pres-
sure, as read from the water manometer and a barom-
eter, was reduced to millimeters of mercury at 0° C.
PROBABLE ERRORS
The sources and magnitudes of the errors in the
vapor pressures of potassium chloride determined by
this method may be classified as follows:
(1) Errors in measuring and controlling the temperature in
the vapor pressure tube. It is believed that the temperature in
the vapor pressure tube was determined correctly within ±10°,
and that the variation of temperature during the course of an
experiment was well within these limits. The maximum magni-
tude of these errors, therefore, varies from 9 per cent at 1044 °,
where a change of 10° in temperature makes a difference of 2.14
mm. in a total vapor pressure of 24.1 mm. of mercury, to 13
per cent at 801°, where a variation of 10° changes the vapor
pressure 0.205 mm. in a total of 1.54 mm.
(2) Errors in determining the volume of gas passed through
the vapor pressure tube. These errors may be due to (a) leaks
in the system, (b) changes in temperature and pressure during
the experiment, (c) inaccuracy in finding the amount of water
displaced. The errors due to leaks in the system were carefully
guarded against and are believed to be absent or at least negli-
gible. Those due to changes in temperature of the gas in the
measuring apparatus were always less than 0.5 per cent, and
those due to changes in pressure not more than 0.2 per cent.
The error in weighing the water displaced was 0.1 per cent, or
less.
(3) Errors in determining the amount of potassium chloride
volatilized varied from less than 0.2 per cent at 1044°, where
the error was not more than 0.1 or 0.2 mg. in weighing and the
amount lost by volatilization was from 110.0 to 131.3 mg., to
3 or 4 per cent at 801°, where the amount volatilized was 5.4 to
7.8 mg.
(4) Errors due to volatilization of the potassium compound
while the boat was being placed in and removed from the tube.
This error was never greater than the error in weighing, for
whenever it was evident that a weighable amount of the potas-
sium salt was being lost in this manner the amount was
found by blank determinations and a correction applied. Hence
this error is included in the errors in weighing.
(5) Excess volatilization of the potassium compound due to
back diffusion of the vapor against the gas stream and condensa-
tion on cooler portions of the tube and plug in front of the vapor
pressure chamber. The magnitude of this error is hard to esti-
mate. It was kept small by having the loosely fitting plug fit
as tightly as possible and still allow for rapid removal and re-
placement, and by increasing the velocity of the gas stream
whenever it became evident that the back diffusion was causing
material error. It is this error which limits the application
of the method to vapor pressures under 25 or 30 mm., on account
of the difficulty of working with gas-stream speeds above 200
cc. per minute. It is believed that the amount of this error is
never greater than the extreme variation of a single determina-
tion from the mean straight line used in extrapolating, whjch
is never over 5 per cent.
(6) Low volatilization due to partial saturation of the gas
with potassium compounds volatilized from condensations in
the tube during previous experiments. To avoid this error as
far as possible, air was passed through the tube for some time
between experiments. If allowed to accumulate, these condensa-
tions became a serious source of error, and when it became
evident that they were seriously interfering, the tube was flushed
out with air while heated at a temperature considerably higher
than that at which the experiments were to be run, or else a
new tube and new plugs were used. Owing to these precautions
and the fact that this error is somewhat compensated for by the
back diffusion mentioned in (5), it is thought that the magni-
tude of this error is never over 5 per cent.
(7) Errors due to uneven distribution of the vapor of the
potassium compound in the gas stream over the boat. The
direction and magnitude of these errors is difficult to estimate.
Their presence was shown in some of the preliminary work on
potassium chloride, where it was found impossible to get dupli-
cates that checked using two different platinum boats, one of
which happened to be deeper and narrower at the top than the
other. The results using the narrow boat were invariably lower
than those with the wider boat, due to the fact that a pocket of
stagnant saturated gas was formed in the top of the narrow
boat and hindered evaporation of the potassium compound.
In the determinations reported, shallow wide boats were used and
closely agreeing duplicates were obtained. It is believed that
under these conditions the errors of this class are not serious.
(S) Errors due to reaction of the potassium chloride vapors
with the impervite tube and plugs. Undoubtedly there was
some reaction between the vapors and the material of which the
tube and plugs were made, and this would tend to absorb the
potassium chloride vapors and give high results. However, on
account of the rapidity of the gas stream and the very small
amount of vapor present in the gas, it is thought that the error
due to this cause is entirely negligible.
(9) Errors in extrapolation. The partial pressures were
plotted against the speeds of the gas stream on coordinate paper,
and the straight line which agreed with the greatest number of
points was extended to zero speed. To check the accuracy
of this graphic method, the equations for the lines through pairs
of mean results for different speeds were written and solved
for the pressure (x) at zero speed (y = 0). The mean of the
pressures thus found, which agreed very closely with the pres-
sure found by the graphic method, was taken as the vapor pres-
sure at the temperature in question. The extreme variation of
the pressure values thus calculated from the mean value was
about * 10 per cent, and it is believed that the vapor pressures
here reported are reliable within these limits.
VAPOR PRESSURE OF POTASSIUM CHLORIDE
It has been shown by Nernst7 that the vapor density
of potassium chloride corresponds to the simple for-
mula KC1. Hence in determining the vapor pressure
of this compound the amount volatilized can be found
directly by loss of weight. The salt used was from a
2-lb. bottle of J. T. Baker Chemical Company's C. P.
Analyzed Potassium Chloride. According to the label
it contained 0.001 per cent or less of each of the follow-
ing impurities: iron, calcium oxide, magnesium oxide,
and sulfuric anhydride, and also a trace of sodium.
Qualitative tests for the above impurities showed that
they were present only in extremely minute quan-
tities. To expel moisture and avoid mechanical loss
from decrepitation, the salt before being weighed for
analysis or for use in a vapor pressure determination
was fused in a weighed platinum boat. The total
potassium present was determined both by the per-
chloric acid method, which separates any sodium
which might be present, and by evaporating a weighed
portion of the fused chloride with an excess of sulfuric
acid in a platinum dish, igniting to constant weight
and weighing as potassium sulfate. The results cal-
culated as potassium chloride by the perchlorate
method were 100.10 and 100.05 per cent, and by the
sulfate method, 99.98 and 99.94 percent. It is safe to
conclude, therefore, that the fused salt is practically
pure KC1. Analyses of the residues from the plat-
14
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
inum boat after vapor pressure determinations showed
that these also were pure potassium chloride. The
potassium chloride left in the boat after Expts. 58 to
63, inclusive, weighed 0.2040 g., and yielded 0.2387 g.
of potassium sulfate, which is equivalent to 0.2042 g.
KC1; the residue from Expts. 67 to 71, weighing 0.5980 g. ,
gave 0.6990 g. of K2S04, equivalent to 0.5981 g. of KC1.
The results of the experiments with potassium chlo-
ride at three temperatures are given in Table II. In
Fig. 4 these results are plotted, using the partial pres-
sures of potassium chloride as abscissas and the speed
of the gas stream in cubic centimeters per minute as
ordinates. The values for the vapor pressures ob-
tained by reading the partial pressures at zero speed
are: 1.54 mm. at 801°, 8.33 mm. at 948°, and 24.1 mm.
at 1044°.
Table II — Vapor Pressure op Potassium Chloride
Nitrogen Partial
Expt. Cc. per Min- Milli- .— KC1 Volatilized^ Pressure
No. Minute utes moles Grams Millimoles Mm. Hg ° C
78.2
77.9
100.1
99.8
108.8
118.6
118.7
119.9
119.7
132.2
132.9
153.0
153.5
184.3
183.0
152.9
154.0
135.9
134.2
80.0
80.5
80.1
77.7
79.4
79.5
80.2
80.2
82.7
77.1
75.1
75.4
82.3
81.6
75.0
82.5
78.8
77.9
0.0074
0.0078
0.0068
0.0065
0.0059
0.0054
0.0054
0.0337
0.0338
0.0318
0.0297
0.0244
0.0239
0. 1146
0.1110
0.1162
0.1268
0.1313
0.1285
0.099
0.105
0.091
0.087
0.079
0.072
0.072
0.452
0.453
0.426
0.398
0.327
0.321
1.54
1.49
1.56
1.70
1.76
1.72
0.93
0.99
0.85
0.82
0.77
0.69
0.69
4.24'
4.25>
3.88
3.89
3.28
3.21
13.9
13.6
15.5
15.3
16.6
16.4
801
803
800
800
802
945
945
948
949
948
947
1040
1046
1045
1042
1044
1046
plotting the line to determine the vapor pressure, the values 4.36
' corresponding to the temperature 948° were used.
VAPOR PRESSURE OF KC1
V
9
48"
C.
^
v
—
—
„>
IM
It
44
"C
-ISO
#
—
•in
—
Millimeters of Mercury
60 80 I M t!0 140 160 30 40 SO 60 70 80 12 14 16 I) 20 Tl 24
Fig. 4
To extend the usefulness of the data obtained, the
vapor pressure curve for potassium chloride from 800°
to 1500°, the boiling point determined by Borgstrom,4
was constructed. Using the values for P found at
801° and 1044°, together with the boiling point, 1500°,
the values of the constants in the empirical and approx-
imate formula of Nernst8
Xo
LogP
+ 1.75 log T-
T + C
4.571 T ' 4.571
were calculated. The simplified formula thus found
for potassium chloride is:
—5326
T
1000 1100 1200 1300 1400
Temperature °C.
Table III — Vapor Pressures
- — Temperatu
•C.
801
948
1044
1100
1150
1200
1250
1300
1350
1400
1450
1500
1 Abs.
1074
1221
1319
1373
1423
1473
1523
1573
1623
1673
1723
1773
' Potassium Chloride :
1500° C.
^— Pressure-
Calculated C
Mm. Hg ]
1.54
9.06
24.1
40.4
62.5
94.4
139.0
202.0
288.0
404.0
558.0
760.0 7
ETWEEN 80(1°
The points on the vapor pressure curve calculated
by this formula are given in Table III. The curve
drawn through these points is shown in Fig. 5.
An approximate value for the latent heat of evap-
oration of potassium chloride can also be calculated
from its vapor pressures by means of the van't Hoff
equation written in the form:9
P, P, X / 1 1 \
Logp?r;-log£T; = 4-571 vt7~t:J
The results of these calculations are given in Table I V .
Table IV — Latent Heat of Evaporation op Potassium Chloridk
Calculated from van't Hoff's Equation
Molecular Heat
Temperati
°C.
801
948
1044
1500
res Pressures
Mm. Hg
1.54
8.33
24. 1
760.0
Mean Value
of Evaporation
X
—27,600
—32,800
—32 000
—30,800
LogP =
+ 1.75 log T + 0.000511 T — 0.7004
VAPOR PRESSURE OF POTASSIUM CARBONATE
Potassium carbonate was the salt used in the first
vapor pressure determinations made because it was
Feb., 192 1
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
115
thought that the conditions of volatilization of potash
from potassium carbonate most nearly approach the
condition of volatilization of potash from a cement
mixture to which no special volatilizing or releasing
reagent has been added. The salt used was a special
grade of chemically pure potassium carbonate. When
kept in a glass-stoppered bottle, which was nearly full
and which was opened only as much as was necessary
in removing the portions used, it did not seem to
change in composition. The portions used for anal-
ysis or in the vapor pressure determinations were
quickly transferred to a platinum boat and this at
once placed in a glass-stoppered weighing bottle. The
sample weighed in this manner gave on analysis by
evaporating in platinum with an excess of sulfuric
acid, heating over a M6ker burner, and weighing as
potassium sulfate, the following results:
> — ■ Per cent .
KiO K2CO3
(a) 65.17 95.61
(i>) 65.31 95.83
(<0 65.19 95.65
(d) 65.21 95.69
Mean 65.22 95.70
The results by the perchlorate method which would
separate any sodium present were:
> Per cent
K2O KsCOj
(e) 65.22 95.70
(.0 65.15 95.59
Mean 65.19 95.65
The sample, therefore, is practically free from sodium,
and qualitative tests showed it to be free from appre-
ciable amounts of other impurities, except moisture and
possibly bicarbonate. On account of the absence of
nonvolatile impurities the amount of potassium oxide
remaining after a vapor pressure determination was
found by dissolving the residue from the platinum boat
in a platinum dish, evaporating with an excess of
sulfuric acid, and weighing the potassium sulfate
formed.
After numerous unsuccessful attempts to obtain
constant weight and constant composition by drying
the salt at temperatures from 120° to 900° C, it was
decided to use the sample as analyzed above. Atten-
tion is therefore called to the fact that the sample
used contained about 4 per cent of moisture, and to
the probability of the results as reported being slightly
higher than the true vapor pressures of anhydrous
potassium carbonate, due to the formation of a small
amount of potassium hydroxide in heating the undried
salt.
To calculate the partial pressure of the vapor of
the potassium salt it is necessary to make an assump-
tion regarding the molecular weight in the vapor state.
In these experiments the amount of carbon dioxide
absorbed by soda lime in the absorbing train agrees
roughly with the amount of potassium oxide lost by
volatilization. It seems probable, therefore, that potas-
sium carbonate on volatilizing decomposes as follows:
K2C03 — > K20 + COa
Hence the vapor pressures were calculated for K20,
using the assumed molecular weight of 94.2. In the
calculations the number of millimoles of carbon di-
oxide was included in the total number of millimoles
whenever the amount of carbon dioxide evolved was
sufficient to affect materially the final results.
The data and results of the experiments at two tem-
peratures are given in Table V, and the plots of the
results giving the vapor pressures at these temperatures
are shown in Fig. 6. The vapor pressures thus ob-
tained are: 1.68 mm. at 970° and 5.0 mm. at 1130° C.
Table V — Vapor Pressure of Potassium Oxide
Carbonate
Cc. ,— K!0 Lost-^
Expt. per Min- ^Millimoles of — . Milli-
No Min. utes N2 COi HiO Grams moles
n Potassium
Partial
Pressure
of EiO
Mm. Hg °C.
4 78 23 79.4 0.1 1.0 0.0068 0.072 0.68 970
5 79 22 76.8 0.1 0.9 0.0083 0.088 0.86 970
6 51 37 83.6 0.1 1.5 0.0109 0.116 1.03 970
7 51 36 81.6 0.1 1.1 0.0103 0.109 1.01 970
10 35 50 77.3 0.1 1.0 0.0119 0.126 1.21 970
11 35 50 77.7 0.1 0.7 0.0122 0.130 1.25 970
15 51 35 78.4 0.5 0.9 0.0390 0.414 3.9 1130
16 51 35 78.4 0.5 1.1 0.0471 0.500 4.7 1130
17 80 23 80.0 0.4 1.1 0.0311 0.330 3.1 1130
18 80 22 77.7 0.4 1.0 0.0309 0.328 3.1 1130
19 102 16 71.8 0.4 1.1 0.0243 0.258 2.7 1130
THE VAPOR. PRESSURE OF K,0 IN K, CO,
970°C
\
II30°C.
100
\
\
\
100
80
\
\
\
\
80 ^
35
S
\
i
.$ 60
v
\
en §
s.
\
b 40
\
\
AT, cT
1
40
5.
\
or, "-->
<o ia
\
20 (j
\
\
0.6 1.0 1.4 1 1.8 :o ! 40 1 6.0 1 1
OS 1.2 16 30 5.0
M///imeters /ig.
Fig. 6
VAPOR PRESSURE OF POTASSIUM SULFATE
On account of the impossibility of obtaining correct
results in the determination of either the potassium
or the sulfate radical in potassium sulfate by the or-
dinary methods of quantitative analysis, the salt used
in these vapor pressure measurements was prepared by
treating some of the same potassium chloride as was
used in the vapor pressure determinations of that salt
with pure sulfuric acid in a platinum dish, and heat-
ing the resulting potassium sulfate over a M6ker burner
to constant weight. Since this temperature was
not high enough to melt the potassium sulfate, before
using it in a determination it was melted in a platinum
boat by being placed for 2 or 3 min. in the vapor pres-
sure tube. It was found that no loss of weight re-
sulted. An examination of the residue after a series
of vapor pressure determinations by evaporating it in
platinum with an excess of sulfuric acid and heating
to constant weight showed that the residue also was
pure potassium sulfate. Hence as there was no evi-
dence of dissociation on heating and since the vapor
density of potassium sulfate has never been deter-
mined, the assumption was made that the vapor cor-
responds to the formula K2S04, molecular weight 174.4.
The partial pressures of potassium sulfate were cal-
culated on the basis of this assumption.
116
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
Table VI — Vapor Pressure op Potassium Sulfate
Milli-
-— KsSOi Lost->
Partial
Min-
moles
Milli-
Milli-
Pressure
utes
Ni
grams
moles
Mm. Hg
0 C.
48
79.4
5.2
0.030
0.29
1129
48
79.6
6.7
0.038
0.36
1129
38
78.0
4.9
0.028
0.27
1127
27
78.5
3.5
0.020
0.19
1129
23
78.1
2.6
0.015
0.15
1127
23
77.5
2.5
0.014
0.14
1126
38
82.6
4.1
0.024
0.22
1131
The results of the experiments with potassium sul-
fate are given in Table VI, and the plot showing the
vapor pressure is given in Fig. 7.
THE VAPOR PRESSURE OF K0SO4
1130° C
100— - *"' - — 100
0 0.10 Q20 0.30 040 050 0.60
Millimeters Hg.
Fig. 7
VAPOR PRESSURE OF POTASSIUM HYDROXIDE
An exact determination of the vapor pressure of
potassium hydroxide presents many difficulties on ac-
count of the extreme chemical activity of this com-
pound. First, it is difficult to prepare a 100 per cent
pure sample to use, and it is perhaps even more diffi-
cult to preserve it and to handle it for use in the ex-
periments. It is also quite a problem to find a con-
tainer made of material which is not attacked by the
hot liquid, and of such shape that it will allow free
evaporation and at the same time prevent loss of the
liquid, which shows an unusual tendency to creep out
of the container. Again there is undoubtedly some
action between the vapors and the walls of the tube
and ends of the plugs in the apparatus, and finally
the composition and molecular weight of the vapor
is not known. In view of the other uncertainties it
did not seem to be worth while to spend a large amount
of time preparing a special grade of pure hydroxide
for the determinations, and it was thought that results
which would give much light on the question of the
commercial volatilization of potash could be obtained
by use of a sample of chemically pure potassium hy-
droxide from a reliable dealer in chemicals. The ma-
terial used, therefore, was from a newly opened bottle
of chemically pure potassium hydroxide, purified by
alcohol and cast into sticks. A stick of this material
was rapidly crushed in a mortar into pieces weighing
from 0.3 to 0.6 g., and these pieces were quickly placed
in separate glass-stoppered weighing bottles and
weighed as soon as possible. Some of the weighed
pieces were used in the vapor pressure determinations
and others were analyzed. The analyses by the per-
chloric acid method gave for the total potassium cal-
culated as hydroxide: 84.67, 84.35, 84.80, 84.45, and
83.98, an average of 84.45 per cent for all of the de-
terminations made. The main impurities, water and
carbonic acid, should not materially interfere with
the volatilization.
In solving the question of containers, both platinum
and nickel were tried before silver was finally selected.
In the final experiments a weighed piece of potassium
hydroxide was contained in a boat of pure silver foil.
This inner silver boat was placed in an outer boat
also of silver foil, and slightly longer, wider, and shal-
lower. The outer boat in turn was set into a larger
nickel boat which served as a support in placing the
charge in and removing it from the vapor pressure
tube. The object of the outer silver boat was to catch
the liquid potassium hydroxide which creeps over the
sides of the inner silver boat and thus prevent its loss
or its action on the nickel boat. This it did success-
fully, for in no case was there evidence that the liquid
had reached the outside of the second silver boat.
The upper edges of the nickel boat after an experiment
were found slightly attacked, evidently by the vapors,
which formed a little dark, greenish gray powder. The
residue in the silver boats was almost colorless to light
gray, effervesced only very slightly with water, and gave
no odor of free chlorine when the water solution was
made acid with hydrochloric acid. The silver of the
two boats after removal of the residue with water and
hydrochloric acid was bright and showed no evidence
of having been attacked. The hydrochloric acid solu-
tion was perfectly clear, proving that no silver had
gone into solution. This hydrochloric acid solution
was evaporated with an excess of perchloric acid, and
the total potassium weighed as potassium perchlorate
and calculated to potassium hydroxide. The loss of
potassium hydroxide by volatilization was then ob-
tained by difference.
Since the formula and molecular weight of the vapors
at the temperature of the experiments were not known,
it was necessary to assume a molecular weight for the
vapors in order to calculate the results as partial pres-
sures. The statement of Roscoe and Schorlemmer,10
evidently based upon the work of Deville, that the
vapors of potassium hydroxide decompose at a white
heat into potassium, hydrogen, and oxygen, needs qual-
ifying, for this decomposition, according to Deville's
own report,11 takes place in the presence of incandes-
cent iron. Moreover, according to Deville in the same
report, the decomposition ceases if the temperature is
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
I [',
lowered below a white heat. Further, according to
Watts,12 who does not give the authority for the state-
ment, potassium hydroxide when heated alone does not
decompose at any temperature. Since the tempera-
ture of the experiments here reported, 795° C, is far
below a white heat, it is not probable that dissociation
takes place to an appreciable extent. Hence it was
most simple and seemed entirely justifiable to assume
that the vapors given off were KOH with a molecular
weight of 56.1.
THE VAPOR PRESSURE OF KOH AT 795°C.
The data of the experiments, together with the re-
sults calculated on this basis, are given in Table VII,
Table VII— Vapor Pressure of
Potassium Hydroxide
Cc.
Milli-
KOH Volatilized
Partial
Jxpt
per
Min-
moles
Milli-
Milli-
Pressure
No.
Min.
utes
Nt
grams
moles
Mm. Hg
" C
127
182
10
81.2
32.5
0.58
5.4
794
129
158
11
77.9
27.2
0.48
4.6
795
180
158
11
77.8
31.4
0.56
5.4
790
LSI
151
12
80.8
35.3
0.63
5.9
793
IK-'
150
12
80.4
34.4
0.61
5.7
795
lK.f
133
13
77.4
34.7
0.62
6.0
794
184
133
13
77.2
41.3
0.73
7.1
795
185
121
15
81.2
44.0
0.78
7.2
794
186
118
15
78.7
38.5
0.69
6.6
795
and these results are plotted and extrapolated in Fig.
8. On account of the possibility of variation in the
composition of the pieces of the sample used in the
different experiments, which variation probably ex-
plains the fact that three of the nine points are at slight
variance with the mean straight line, a high degree of
accuracy is not claimed for the vapor pressure found,
namely, 8 mm. at 795° C. It is believed, however,
that this result is not in error more than 25 per cent,
and that the result plainly shows that the vapor pres-
sure of potassium hydroxide at 800° C. is almost as
large as that of potassium chloride at 950° C, and con-
siderably larger than the vapor pressure of potassium
oxide in potassium carbonate at 1130° C.
VAPOR PRESSURE OF POTASSIUM OXIDE IN NATURAL
SILICATES
In the attempt to determine the vapor pressure of
potassium oxide in natural silicates, three samples were
used, each of which was ground in agate to pass a 200-
mesh sieve.
(1) Leucite — This consisted of portions of two large
tetragonal trisoctahedron crystals. The original crys-
tals were about 0.75 in. in diameter, colored gray on
the outside, and glassy, almost transparent, inside. The
sample after grinding was pure white, and analyzed
19.05, 19.10, 18.97, and 19.04; mean, 19.04 per cent
K20.
(2) Feldspar — This sample was part of a crystal of
orthoclase, with angles of 90°, very light gray in color,
with a slight tinge of red and a glassy luster. The
powder was almost pure white with a slight gray tint.
Duplicate analyses gave 13.90 and 13.97 per cent of
K20.
(3) Glauconite — The sample was furnished by the
Coplay Cement Company. According to their anal-
ysis it contained:
Silica 40 . 56
Alumina and ferric oxide 30.40
Calcium oxide 9 . 58
Magnesium oxide 2 . 09
Potassium oxide 6.06
Loss on ignition 10.52
It was found to contain iron equivalent to 20.85 per
cent of ferric oxide, and analysis gave 6.10, 6.04, 6.07,
6.03, and 6.00; mean, 6.05 per cent of K20.
In the experiments a weighed portion of about 0.5
g. was heated for 48 min. in a platinum boat in the
vapor pressure tube, while dry nitrogen was passed
through at a speed of 35 to 37 cc. per minute. Within
the limit of accuracy of the analyses (about 0.0005 g.
of K20 in a 0.5 g. sample) there was no loss of potas-
sium in any of the silicates at 1335° C. or lower.
Hence the vapor pressure of potassium oxide in these
three natural silicates when heated alone at temper-
atures under 1350° C. is entirely negligible.
The results of experiments with the three silicates,
showing loss of weight and change of state at three
temperatures, are given in Table VIII.
Table VIII — Results of Heating Potassium-Bearing Silicates for
48 to 50 Min.
Loss of
Weight
Expt. Temp. Silicate Per Loss of Residue,
No. ° C. Used cent KiO Appearance, etc.
25 1130 Leucite 0.60 None White, no sintering
28 1245 Leucite 0.73 None White, trace of sintering
32 1335 Leucite 0.74 None White, slightly sintered
24 1130 Feldspar 0.00 None Pale gray, no sintering
27 1245 Feldspar 0.04 None Pale gray, slightly sintered
31 1335 Feldspar 0.08 None Nearly all fused to a
colorless glass
23 1130 Glauconite 11.59 None Reddish brown, sintered
29 1245 Glauconite 12.13 None Dark red, fused
30 1335 Glauconite 12.47 None Dark greenish glass
SUMMARY
I — The vapor pressure method of von Wartenberg
has been adapted to the study of the vapor pressures
of potassium compounds and the vapor pressures shown
in the following table have been determined.
118
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
Vapor Pressures i
° C. Hydroxide Chloride
795 8
801
948
970
1044
1130
1335
II — From the results of the vapor pressure measure-
ments with potassium chloride at 801° and 1044° C,
together with the boiling point of this compound as
given by Borgstrom, the Nernst vapor pressure for-
mula for potassium chloride has been calculated to be:
—5326
Log P = — f- 1.75 log T + 0.000511 T —0.7064
By means of this formula the vapor pressure curve for
potassium chloride from 800° to 1500° C. has been
constructed.
Ill — It has been established by these vapor pressure
measurements that the order of volatility of those
potassium compounds which are most important in
the recovery of potash from cement or other silicate
mixtures is as follows: Hydroxide, chloride, oxide
from carbonate, sulfate, and natural silicates.
REFERENCES
I— J. Ind. Eng. Chem., 9 (1917). 253.
2— Ann., 138, 263; Jahresb., 1866, 770.
3—/. Am. Chem. Soc, 19 (1897), 155.
4 — Med. Finska Kemistsamfundet (Swedish), 24 (1915), 2; through
Chem. Abs., 9 (1915), 2361.
5— Z. anorg. Chem., 85, 234; J. Am. Chem. Soc, 36 (1913), 1693.
6— Z. Elektrochem., 19 (1913), 482; Z. anorg. Chem., 79 (1912). 76.
7—Nachr. kgl. Ges. Gottingen, 1903, 75; through Zenlr., 1903, Vol. II,
17.
8 — Nernst. W., "Theoretical Chemistry." 1911, p. 719.
9—Z. Elektrochem.. 19 (1913), 484.
10 — Roscoe and Schorlemmer. "Treatise on Chemistry," Vol II,
"The Metals," 1907, p. 321.
11— Compl. rend., 16 (1857), 857.
12 — Watts, "Dictionary of Chemistry," 1868, Vol. IV. p. 702
RUBBER ENERGY1
By Wm. B. Wiegand
Rubber Section, Ames Holden McCready, Ltd. , Montreal. Canada
It is proposed to discuss very briefly and nonmath-
ematically some of the many interesting energy rela-
tionships of vulcanized rubber.
ENERGY STORAGE CAPACITY
In the accompanying table is shown what is known
as the "proof resilience" of the chief structural ma-
terials. This is defined as the number of foot pounds
of energy stored in each pound of the material when
it is stretched to its elastic limit. You will observe
that tempered spring steel has less than one one-hun-
dredth the resilience of vulcanized rubber, and that
even hickory wood, its nearest rival, also shows less
than one per cent of the resilience of rubber.
This property of course is directly made use of in
aeroplane shock absorbers, etc., but our present ref-
erence to it is made with a view to discussion, first, of
the character of this stored energy and its transforma-
tion into thermal energy of two kinds; and, second,
the modification and in fact remarkable increases in
• Presented before the Rubber Division at the 60th Meeting of the
American Chemical Society, Chicago. 111.. September 6 to 10, 1920
energy storage capacity made possible through the
admixture of suitable compounding ingredients.
Table I — Proof Resilience
Ft. Lbs. per
Material Cu. In.
Gray cast iron 0.373
Extra soft steel 3 . 07
Rail steel 14.1
Tempered spring steel 95.3
Structural nickel steel 14.7
Rolled aluminium 7.56
Phosphor bronze 4 . OjB
Hickory wood 122.5
Rubber 14.600.00
THERMAL EFFECTS
What happens to the mechanical work done on a
rubber sample when it is stretched to any given point?
Is it in the form of potential energy of strain, as in
the case of a steel spring? The answer is, "No."
Has it all been irrecoverably lost in the form of heat.
as when a lump of putty is flattened out? No. Or
lastly, as when a perfect gas is isothermally compressed,
has the work done on the sample been turned into an
equivalent amount of heat which is, however, con-
vertible back into work during retraction? Here
again the answer is, "No."
The fact is that rubber has all three properties com-
bined. When you stretch a rubber band, some of the
energy is stored as potential energy of strain, exactly
as when you stretch a steel spring. Another fraction
of the energy input is turned into what may be called
reversible heat. You can easily feel this heat on
stretching a rubber thread and touching it to your
lips. The rest of the energy input or work done on
the rubber appears in the form of frictional heat.
RETRACTION
We will suppose that the extension was made rap-
idly (». e., adiabatically) and consider what happens
on the retraction journey, which we will assume to
take place rapidly and immediately. First of all, the
potential energy of strain will nearly all be returned
in the form of useful work, exactly as in the case of
the steel spring. Secondly, the reversible heat which
on the outward journey acted to increase the tem-
perature of the sample will be re-absorbed, transformed
into useful work, and therefore cause no energy loss.
Finally, the frictional heat developed during extension
will be increased by a further amount on retraction,
at the expense of the potential energy of the stretched
sample.
Thus, when the rubber has been stretched and al-
lowed to return to substantially its original length, it
will differ from its original state only by the total
amount of frictional heat developed. By the law of
conservation of energy, we can at once say that this
frictional heat is exactly represented by the difference
between the mechanical energy input and output of
our system. This phenomenon is, of course, known
as hysteresis, and is exhibited by all structural mate-
rials. The fact that in the case of rubber the energy
storage capacity is several hundred times greater than
in the case, say, of steel, explains why hysteresis phe-
nomena become relatively of such cardinal importance
to rubber technologists.
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
119
REVERSIBLE HEAT AND THE JOULE EFFECT
Suppose we extend a rubber sample and allow the
reversible heat thus generated to disappear. In other
words, we stretch it isothermally. We are then deal-
ing with a system substantially in equilibrium. The
two factors governing this equilibrium are, first, the
load on the rubber, and, second, the thermal condition.
Any change in the equilibrium requires a change in
these two factors. Conversely, a change in either of
these factors will shift the equilibrium. Now one of
the fundamental properties of any equilibrium is that
when any factor is changed the equilibrium will be
shifted in such a way as to offset the change in the
factor. Thus, if the load is increased, the sample will
stretch and become stiffer so as to resist the increased
load. Similarly, if the temperature of the sample is
increased, the rubber will contract, since in so doing
heat is used up and in this way the disturbance min-
imized.
This contraction on heating was predicted by Lord
Kelvin, after Joule had discovered, or rather redis-
covered, the development of heat during extension.
Metals and most other rigid bodies behave, of course,
in a totally different fashion. Instead of generating
heat on extension they consume heat and become
cooler, with the result that the application of heat to
a stretched metal wire causes it to expand instead of
contract, as in the case of rubber.
The Joule effect has been subjected to many misin-
terpretations, such, for example, as attributing it to a
huge negative temperature coefficient of expansion,
which is, of course, incorrect, since rubber has in fact
a large positive coefficient. Others have attempted
to explain the phenomenon by assuming an increase
in Young's modulus. Bouasse, the French investigator,
who has done sdch masterly work on the elastic
properties of rubber, disproved this hypothesis, how-
ever, and showed in fact that Young's modulus grew
smaller with increased temperature.
The writer has not done any experimental work on
the reversible heat which governs the Joule effect,
but there can be no doubt as to its technical impor-
tance. Thus, for example, the internal state of a solid
tire tread as well as breaker conditions in large pneu-
matics is clearly bound up with the reversible thermal
effect as well as with the frictional thermal effect.
Every time the tire tread impacts upon the road sur-
face each part of the rubber stock traverses a stress-
strain cycle. Even if we admit that the reversible
heat generated during extension is reabsorbed during
contraction, we have to consider the gradual building
up of internal temperatures due to accumulation of
frictional heat. This increase in temperature, acting
through the Joule effect, will lessen the extensibility
of the heated rubber as compared with adjacent re-
gions at lower temperatures, thus setting up strains
which doubtless play a role in breaker separation, the
bane of large-size pneumatics. It is therefore highly
desirable to work out rubber compounds which will
develop not only minimum frictional heat, but also
minimum reversible heat. Quantitative measurements
of the Joule effect with different compounds and
different cures would serve as an index to this quan-
tity.
MECHANICAL PICTURE OF RUBBER
The diagram in Fig. 1, which was first suggested by
a former colleague, Dr. F. M. G. Johnson, of McGill.
helps clarify one's mental picture of the thermody-
namical phenomena associated with rubber strains.
Fig. 1 — Mechanical Picture of Rubber
Rubber may be viewed as a combination of a cylinder
of gas, a steel spring, and a friction member. Follow-
ing this picture, extension of the rubber is accompanied
in the first instance by compression of the gas, thus
generating the reversible heat, Qr. In the second
place, the steel spring is compressed, thus generating
the increase in potential energy of strain, E. Lastly,
the friction element operates through the extension,
generating nonreversible heat, Qf. When the rubber
retracts, the gas expands, the spring retracts, and the
friction element contributes another increment to the
nonreversible heat.
Suppose now the sample is extended and we apply
heat to the system. The gas in the chamber will ex-
pand so as to use up heat, raising the weight W, thus
shortening the rubber and so constituting the Joule
effect.
FRICTIONAL HEAT OR HYSTERESIS
Although the reversible heat has doubtless a decided
technical significance, by far the most important
energy transformation is that of useful work into heat
through hysteresis, and a short account will now be
IS
TEE JOURS AL OF IXDUSTRIAL AXD EXGIXEERIXG CHEMISTRY Vol. 1
carried out under the
ippel.
:"nod consisted in ger.
recording
:3n up to vai
tions.
sis loop was readings cal-
:t pounds of ene- I to one cubic
In order to obviate the ir.::
: ensile machines, and for other reasons of
a special machine was devised, the prin-
cipal features of which were the alignment of a helical
steel spring with the sample and the use of extremely
light and nicely fitting parts. The rubber sample was
a standard test piece about 0.1 in. in thick-
ness, 0.25 2 en shoulders. The
ends of the test piece were secured in special light
weight clamps designed practically entirely to obviate
creeping. The spring extension measured fr-
aud the separation of the clamps, the strains.
Through the use of this special machine it was pos-
sible to generate stress-strain cycles both under rapid,
or adiabatic, a*d slow, or isothermal, conditions.
ISOTHERMAL CYCLES ADOPTED It is of COUTSe ob-
vious that the size and char the hysteresis
cycles will depend on whether they are generated
.callyoris:" Under the former con-
ditions, the c iriational heat developed
on the extension journey are only slightly dissipated.
and so act to incr: aness of the sample and
alter the trend of the curves. Owing to the difficulties
was not found possible to generate adia-
batic loops at speeds sufficient to allow of concordant
results. The method finally adopted was to generate
the cycles at low speed? le, 20 in. per min-
rmal conditions.
preliminary extensions — It is of course well
known that the area of 1 loop is
i so on. In most
cases, however, the third loop differs only very slightly
from the succeeding loops, and so in our work when
it was the intention to generate the hysteresis loop up
to an elongation of 300 per cei I piece which
had not been otherwise handled after cutting from
the molded slab was put through two preliminary
cycles up to 300 per cent, and then clamped into the
machine, an a -is loop graphically recorded.
In taking a succession of loops at increasing elonga-
tions the same test piece was used and two preliminary
loops made at each elongation. The initial length
upon which the cycles were based was the length
measured a:: preliminary extensions had
eer. raaae.
e or compounds used — The experimental re-
-
compounds used in tire construction. They thus in-
cluded practically -. :tion compounds, lightly
loaded breaker compounds, and more heavily loaded
tread stock. The; naixed in the
ander standard conditions, and given laboratory
I
each case up to cures 275 per cent over the optimum
in some cases.
Hysteresis loops were generated at elongations rang-
ing from 100 to 500 per cent. There is considerable
. :e in opinion as to whether in measuring hys-
teresis one should work toward reaching a fixed per-
centage of the breaking load, irrespective of the elonga-
tion, or work to a definite elongation, irrespective of
the load required. The latter method seems to the
writer the only correct one from the technical stand-
point, in view of the fact that the strains incurred, for
example, by the skim coat, breaker, and tread of a
pneumatic tire are arbitrarily fixed by the inflation
pressure and the load.
RELATION BETWEEN HYSTERESIS LOSS AND CYCLIC
elongation — Fig. 2 illustrates the results obtained
with a typical pure gum, high-grade tire friction with
a breaking elongation of upwards of 900 per cent. This
particular compound contained 5 lbs. of sulfur to 100
lbs. of rubber, of which 60 were first latex rubber and
the other 40 a soft-cured wild rubber. The or.',
ingredients were a small percentage of thiocarbanilide
and 5 lbs. of zinc as activator. The energy units are
expressed as one-hundredths of a foot pound calcu-
lated to a cubic inch of rubber. The relationship is
of the character of a rectangular hyperbola, and the
hysteresis increases very sharply for elongations ex-
ceeding 300 per cent. Viewing hysteresis as frictional
is natural to expect sharply increased friction
to accompany the rapidly increasing lateral compres-
sions in the test piece. Following our mechanical pic-
ture, it is analogous to contraction of the friction
element upon the moving arm.
CYCLIC
1
ELONGATION
VS.
HYSTERESIS LOSS
-
-
1
y
: ELONGATK
Fig. 2
adopiion of standard loop — For comparison of
at compounds and for different cures it was
decided to adopt a standard cyclic elongation, and in
" " reduce experimental error it was of course
desirable to select an elongation lower than 300 per
- lying on the flat portion of the curve. For
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
121
o
J 300
va
at 200
tfl 150
UN
s8er
CURE
VS.
HYSTERESIS
OPT
1QP r-r.iso
OVER
PERCENT CURE
higher elongations the energy loss changes so rapidly
with slight changes in the elongation as to make con-
cordant results difficult. Moreover, a brief calcula-
tion of the strains set up, for example, in the skim coat
of a pneumatic casing run under service conditions
shows that under conditions of standard factory prac-
tice the rubber is strained to an elongation of not much
more than 200 per cent each time the tire flattens
against the road. For comparative purposes we there-
fore adopted a standard cycle of 200 per cent elongation.
RELATION BETWEEN STATE OF CURE AND HYSTERESIS
LOSS
It is commonly held by tire technologists that the
state of cure of the friction and skim coat of the car-
cass has a lot to do with the early or late occurrence of
ply separation.
Fig. 3 does in fact show that the state of cure has an
influence on hysteresis. What is shown as the normal
cure on this chart is the optimum cure as determined
by the tensile product. An under-cure of 50 per cent,
for example, means that if the optimum curing time
is 90 min. at 40 lbs. of steam pressure, the sample was
cured for 45 min. Similarly with over-cures. Curves
A and B are typical skim coat compounds. Curve C
is a breaker compound. It will be observed that min-
imum hysteresis occurs in the over-cured region. It
must, of course, be kept in mind that these data apply
only to cycles of 200 per cent elongation, whereas the
rubber stock in question has an ultimate elongation
of over 900 per cent. Attention must also be called
to the danger of assuming that a slight over-cure is
therefore desirable. Aging conditions must be taken
into consideration, and the writer is of the personal
opinion that the optimum cure or, in many cases, an
even shorter cure is the correct condition. It is also
noteworthy that the actual magnitude of the hys-
teresis values characteristic of high-grade, pure gum
frictions is very low, and that we must look elsewhere
for the true cause of ply separation.
lOOO
900
800
700
1 1
VOLUME OF FILLER
VS.
HYSTERESIS
~B JO R~
VOLS. OF ACTIVE PIGMENT
20 25
MIXED WITH
100 VOLS OF
Fig. 4
RUBBER
THE EFFECT OF COMPOUNDING INGREDIENTS
This presents an enormous field of research, and
reference will be confined to a brief outline of the
basic facts.
Fig. 4 shows hysteresis plotted against the volume
percentage of active pigment associated with 100 parts
of rubber. The first point on the curve shows a pure
gum compound, the second, a lightly loaded breaker
compound containing about 4.5 parts by volume of
active pigment. The third point represents a very
high-grade tread compound containing about 15 vol-
umes of active pigment: the last, another tread stock
containing nearly 24 volumes. By active pigment is
meant a pigment which definitely increases the energy
storage capacity of the compound and includes pig-
ments such as carbon black, lampblack, zinc oxide,
the finer clays, etc. It will be noted that for the par-
ticular stocks used there is a linear relationship be-
tween the amount of hysteresis and the amount of such
pigment present. It is also important to note that
the effect of the addition of a highly dispersed phase
upon hysteresis is much greater than moderate changes
in the state of cure of a compound. It is unnecessary
to emphasize the importance of this result from the
standpoint of practical compounding.
Here again, however, one must use caution not to
overlook the importance of heat conductivity, and it
is entirely within the realm of possibility that a pig-
ment, although markedly increasing the hysteresis and
so also the frictional heat, may at the same time com-
pensate for this by a greatly enhanced heat conduc-
tivity. Thus, for example, carbon black not only
causes high frictional heats, but is also a bad conduc-
tor, whereas zinc oxide, although producing similarly
high hysteresis values, has a very much better heat
conductance.
It may be of some interest to indicate roughly the
actual percentages of energy which are degraded into
heat in these various types of rubber compounds. A
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
pure gum friction or skim coat stock when led through
a hysteresis loop to an elongation of 200 per cent de-
grades about 4 per cent of the total energy into heat.
TIRE PENDULUM
A stock containing about 5 volumes of zinc oxide de-
grades about 8 per cent, whereas a tread stock con-
taining, say, 20 volumes of zinc oxide degrades in the
neighborhood of 14 per cent of the total energy input
in each cycle.
FABRIC ENERGY LOSSES
We have dealt thus far with the degradation of en-
ergy into frictional losses in and by the rubber sub-
stance itself. These are of paramount importance in
the case of solid tires, for example. However, in the
case of pneumatic tires, which consist primarily of
layers of fabric held together and waterproofed by
rubber, we have to consider the extent to which fric-
tional heat is developed by the carcass fabric itself.
It is true that the hysteresis loss of an inflated casing
taken as a whole can be accurately determined by the
electric dynamometer. This, however, is an expensive
machine, and has the further disadvantage of not
being able to determine in what proportion the various
constituent parts of the casing contribute to the in-
tegral result. The writer has therefore applied the
principle of the damped pendulum to the study of
casing energy losses. Briefly, the method consists in
inserting a 1-in. carcass section in the arm of a pen-
dulum which is allowed to swing from a fixed position
until it comes to rest. The more perfectly resilient
the carcass wall, the longer will such a pendulum
swing. In order to analyze the elastic properties of
the various structural components of the carcass, it
is necessary merely to strip off the tread and breaker
and repeat the series of vibrations with the carcass
alone. In order to ascertain the effect of the number
of plies of fabric the carcass is stripped down ply by
ply and the total period of the pendulum redetermined
in each case.
Fig. 5 shows the simplicity of the set-up. The
inch section is gripped by two clamps, the upper one
rigidly fastened to the wall, the lower attached to the
pendulum arm, consisting of thick piano wire about
2 ft. long, weighted down by a cylindrical bob of con-
venient mass, say 0.5 lb. Time will not permit de-
scription of the minute experimental details, some of
which are of considerable importance to the accuracy
of the results obtained, but, briefly, the practice was to
start the pendulum from a position, say, 60° from the
vertical, and take shadow readings on an arc back-
ground by means of a fine needle axially inserted in
the bob. The "total period" of the pendulum is the
number of seconds required for the amplitude to fall
from the fixed arbitrary value, say, when the shadow
of the needle reaches the point C until the shadow
reaches the point D, which is preferably a small dis-
tance removed from the position of rest. The length
of the carcass strip between the clamps may be varied
at will, but is preferably about 2 in.
significance of total period — The total period,
viz., the time required for the pendulum to damp down
from the position C to the position D is clearly a mea-
sure of the time required for the potential energy of
the pendulum system to fall from that corresponding
to the height of its center of gravity when the pointer
is at C to that corresponding to D. It is therefore in-
versely proportional to the rate of generation of fric-
tional heat through the various internal energy losses
in the casing section. If the tire were of theoretically
perfect resilience the pendulum would keep on swing-
ing forever, except, of course, for external losses due to-
air resistance, etc.
A typical series of determinations will serve to fix
our ideas. A 3.5-in. plain casing gave a total period
of 6 min. 42 sec. After removing the band ply of the
carcass, the period increased to 7 min. 37 sec; after
*u
1 1 1
TIRE PENDULUM
m
TP-I
V1^
20
ir>
10
o
1
s
> ;
J
1-
) t
> <
r a
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
123
removing the second ply, to 8 min.; after removing the
third ply, to 10 min. 55 sec. When all the carcass plies
had been removed and the tread and breaker inserted,
the pendulum swung for 21 min. 4 sec. As a matter of
fact, it was found in many hundreds of tests that the
total period of the pendulum when plotted against
the number of plies of fabric in the carcass lay on a
smooth curve, shown in Fig. 6.
This curve is of the exponential type, the equation
of which is
TP = K, X K>N,
where TP is the total period, Ki and K5 are empirical
constants, and N is the number of plies of fabric. An
interesting deduction from this curve is that the fric-
tional losses in a casing are not a linear function of
the number of plies of fabric. As a matter of fact,
the total period for a 5-ply carcass bears the same
ratio to that of a 4-ply carcass, as that of a 4-ply
carcass bears to that for a 3-ply carcass. In other
words, as the number of plies of fabric is increased
the frictional heat increases not in arithmetic but in
geometric progression. This constant ratio we have
called the "ply factor," and its value in a typical
square fabric casing lies very close to 0.7 for ranges
of from 2 to 7 plies. If the total period for a 6-ply
section is 100 min., that for a 7-ply section will be 70
min. If there were no fabric friction, this factor would
of course become unity, except for the small losses due
to the skim coat between the plies.
INFLUENCE OF GUM STOCKS ON CASING ENERGY
losses — It was at first thought that the condition of
the skim coat and friction between the plies of fabric
might profoundly influence the casing energy losses,
and a series of tire sections was therefore prepared of
various degrees of under- and over-cure. To our great
surprise the effect of these exaggerated under- and
over-cures upon the total period of swing was entirely
negligible in every case.
effect of tread and breaker — Our results, fur-
thermore, showed that, for example, in the case of a
3.5-in. 4-ply casing, the total period of swing for the
complete section was almost exactly the same as that
for a 4-in. 5-ply casing, stripped of its tread and breaker.
We thus see that the entire tread and breaker of a
casing contribute no more to the energy losses than
does a single ply of carcass fabric.
cord construction — These remarkable results made
it at once desirable to ascertain the effect of cord con-
struction, the advantages of which, from the stand-
point of internal chafing, seemed obvious. Our ex-
periments fully bore out this idea, and in fact we found
that a 5-in. cord carcass swings almost exactly three
times as long as a square fabric carcass of the same
size. Cord fabric is therefore three times as efficient
as a transmitter of energy as square fabric. Our pur-
pose in thus briefly describing the pendulum method
of investigation is not to expound the behavior of the
various structural elements of a casing, but rather to
illustrate the usefulness of a simple, convenient, cheap,
and yet accurate physical apparatus in helping to
solve the pressing problems of our industry.
effect of pigments on energy storage capacity
So much for the transformations of rubber energy
and in particular its degradation into frictional heat
through hysteresis.
Of equal interest, however, is the study of the total
energy storage capacity of vulcanized rubber and the
profound changes in this quantity which can be in-
duced through the. admixture of suitable ingredients.
The experimental details of this work have been pub-
lished elsewhere.1 The fundamental facts are as fol-
lows:
1 — A pure gum stock is totally unsuitable for some of the most
important technical applications of rubber by reason of its
inability to stand abrasive wear.
2 — The addition in suitable amounts of certain compounding
ingredients enormously improves the wear-resisting power of
rubber. Our investigation as to the reasons underlying these
facts naturally began with a quantitative study of the effect of
the various compounding ingredients upon the mechanical
properties of the stock. These properties are very largely
expressed by the stress-strain curve, and on selecting a suitable
basic mix and adding to it regularly spaced increments by
volume of the most important inorganic compounding in-
gredients, it was at once discovered that profound changes in the
character of the stress-strain curve were thereby induced.
These changes may be divided into two classes.
One class comprises merely a foreshortening of the curve.
Thus, for example, the addition to the basic mixing of increasing
percentages by volume of barytes produces a stock which, when
gradually stressed to the failure point, preserves the same values
of elongation and load as in the case of the pure mixing. The
only difference is that failure occurs earlier. In other words,
this pigment simply dilutes or attenuates the mechanical
properties of the mixing. It plays a passive role.
In the other class the stress-strain relationships are pro-
foundly altered. Thus, for example, if glue or zinc oxide or
one of the blacks be added to the basic mix in increasing amount,
the mechanical properties of the resultant vulcanisate show the
following changes:
First, the curvature of the stress-strain curve is diminished
and at suitable pigment concentrations actually disappears.
That is to say, rubber can be so compounded as to display the
same kind of stress-strain relationship as in the case of steel
and the other rigid structural materials, i. e., Hooke's law ob-
tains. Again, certain of these same pigments, if not added in
excessive amounts, produce compounds, the tensile strength of
which at rupture remains undiminished or even increased over
large compounding ranges. In these cases the final elongation
is, however, markedly reduced. In the other cases, although
linear stress-strain relationships are induced, both tensile
strength and elongation fall off more or less equally
It has been thought justifiable in view of these striking
differences in behavior to call pigments of the second class
active pigments and those of the former class inert pigments.
In Table II are brought together, along with the
energy storage capacities which are here designated
the total energy of resilience, the dispersoid charac-
teristics of the pigments in question, and also the in-
crease in total volume of the compounded rubber when
stressed to 200 per cent elongation. These volume in-
creases, for the details of which you are referred to a
recent paper2 by my colleague, Mr. Schippel, prove
i Can. Chem. J.. 1 (1920), 160: see also abstract in India Rubber World,
63 (1920), 18. Both references give curves illustrating the effect of various
pigments on the energy storage capacity of the rubber.
s This Journal, 12 (1920), 33.
124
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
Table II
Displace- Total
ment of Energy Volu
Apparent S. S. of Incre
Pigment Surface Curve Resilience at 200% El.
Carbon black .. . 1,905.000 42 640 1.46
Lampblack 1.524,000 41 480 1.76
China clay 304 , 800 38 405
Red oxide 152.400 29 355 1 9
Zinc oxide 152,400 25 530 0.8
Glue 152,400 23 344
Lithopone 101,600
Whiting 60,390 17 410 4.6
Fossil flour 50,800 14 365 3.5
Barytes 30,480 8 360 13.3
Base
450
beyond any doubt that particularly in the case of the
inert pigments the application of stress causes a par-
tial separation of the pigment from the rubber with
resultant development of vacua at the poles. In the
active pigments, those which show a positive effect
upon the energy storage capacity, this separation from
the rubber matrix is very slight. Column 2, which
gives the sq. in. of surface per cu. in. of pigment, indi-
cates that the extraordinary differences in behavior are
without doubt attributable to differences in surface
energy. When a stock containing one of the active
pigments is stressed to rupture, the energy required
to do so goes partly towards distorting the rubber
phase and partly towards tearing apart the rubber
from the pigment particle.
Again, the fact that in the case of the active pig-
ments the rubber remains more nearly adhesive to each
particle means more uniform stress on the rubber phase,
and so enhanced tensile properties and energy capacity.
Surface energy has, of course, two factors. The
capacity factor is represented by the specific surface,
and it is the variations in this factor which appear to
predominate in the behavior of the various pigments.
The other factor, the intensity factor, which is repre-
sented by the interfacial surface tension, is also doubt-
less of importance, as is shown by the fact that zinc
oxide occupies a somewhat anomalous position in the
energy column. It is, namely, a more active pigment
than would be indicated by its developed surface.
Briefly, any pigment of a degree of subdivision cor-
responding to a surface development of over 150,000
sq. in. per cu. in. may be expected to belong to the
active class. It must of course be remembered that
the activity of a pigment depends entirely upon the
percentage present in the mixing. Maximum activity
is developed for volume percentages lying between 5
and 25. Inert pigments of course develop no activity
no matter how much or how little is added.
THE STRUCTURE OF COMPOUNDED RUBBER
In view of the important role played by surface
energy in the properties of compounded rubber, and
also in view of the recently demonstrated fact of the
physical separation of the constituent particles from
their rubber matrix under conditions of strain, it is
clearly of importance that we should know something
about the spacial distribution of the component par-
ticles of a mixing. Thus, for example, how much
barytes may one add to a compound before the par-
ticles actually touch each other? How far apart are
the particles of zinc oxide in a tread compound con-
taining, say, 20 volumes of this pigment?
These interparticle distances are of theoretical im-
portance, not only for the proper calculation of the
forces acting upon the rubber phase occupying the
interstices, but also in connection with the influence,
if any, of electrostatic charges upon the pigment par-
ticles during mixing.
Let us first assume that sufficient pigment has been
added to cause actual contact between the particles.
Now it is not at all a simple matter to calculate what
percentage must be added to bring about this condi-
tion. The question involves a study of the theory of
piling. Thus, for example, if we fill a quart measure
with marbles, the number we can get into the measure
depends upon the character of the piling which they
assume. If, after laying in the first layer we place suc-
ceeding layers in such a way that each marble lies ver-
tically over and touching the one beneath, we obtain
what is known as cubical or loose piling* If, however,
we shake the marbles down until they lie together as
closely as possible, the piling assumes a totally different
character, known as normal, close, or tetrahedral piling.
This question of cubical or tetrahedral piling is im-
portant in all studies of granular bodies. Thus, for
example, the rigidity of mortar under the trowel, or
the firmness under the foot of the wet sand on the
seashore are both due to the fact that the granules are
in a condition of close or normal piling, the distur-
bance of which by an external force requires an increase
in the over-all volume, which in turn is resisted by the
vacua which tend to be formed.
If a test tube be loosely filled with sand and sub-
sequently gently tapped, the sand will settle down a
considerable distance in the tube. The sand was orig-
inally more or less loosely piled. It was certainly
not piled in the most loose manner possible, namely, cubi-
cally, but occupied some intermediate position. On
gently tapping the tube the particles are freed, and,
attracted downward by the force of gravity, assume a
spacial arrangement more nearly normal or tetrahedral.
THE PILING OF COMPOUNDING INGREDIENTS We have
now to consider what happens when a pigment is
worked into the rubber in a plastic state on our mix
mills. Owing to the high viscosity of the gum the
force of gravity is not free to act as it did in the case
of the sand in the test tube or the marbles in the
quart measure. Taking first a case where so much
pigment is added that the particles are compelled to
touch each other, it is possible to calculate the amount
of pigment required on the assumption, first, that the
particles are arranged cubically or loosely, and, sec-
ond, tetrahedrally or closely.
On the former assumption, irrespective of the size
of the particles (which are, however, assumed to be
uniformly spherical), the amount required would be
52.4 per cent of the total by volume. On the second
assumption, the figure comes out at 74.1 per cent.
Now it is a well-known fact in mill practice that a
compound containing 50 per cent by volume of pig-
ment is almost unmanageable on the mill. We there-
fore deduce that with the customary amount of mill-
ing the pigment particles probably exist in a condition
more closely approximating the loose or cubical piling
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
125
than the close or tetrahedral piling. The writer has,
however, observed that in working with extremely
heavily loaded stocks it is possible, by continued mill-
ing, to bring about a more or less sharply defined in-
crease in plasticity with the possibility of working in
an additional amount of pigment. With due regard
to the breaking down of the rubber owing to this ex-
cessive milling, it still remains highly probable that
the additional mastication has caused a more even
distribution of the rubber phase throughout the mass,
which is equivalent to saying that the particles have
been rearranged to more nearly normal piling. The
writer has in fact succeeded in milling in over 60 per
cent by volume of pigment in this way (*. e., 60 vol-
umes pigment to 40 volumes rubber).
20 JO 40 50 60 70
Fig. 7 — Interparticle Distance vs. Volume Per cent Pigment
SPACIAL ARRANGEMENT WHEN NOT IN CONTACT Fig.
7 shows interparticle distances for percentages of pig-
ment ranging all the way from 0 to 80 per cent. The
ordinate D shows the distance between the particles
referred to their radius as unity. The upper curve
shows conditions when the particles are tetrahedrally
disposed. Under working conditions in the factory
very few compounds contain more than 35 per cent
by volume of pigment. Taking, for example, a typical
tire tread compound containing, say, 20 per cent of
pigment by volume and assuming tetrahedral arrange-
ments, the particles will be distant from each other by
a little over their own radius. Assuming cubical ar-
rangement they would be closer together, namely, dis-
tant by about three-quarters of their radius. This of
course presupposes spherical shape. In actual prac-
tice, the pigment particles are by no means spherical,
but on the average they are more nearly spherical
than of any other definite geometrical shape, and the
error due to assuming sphericity will not be large.
The question as to whether in such cases where the
particles are not in actual contact one ought to as-
sume a tetrahedral or a cubical space arrangement is
(at least to the writer) very difficult to answer by
mathematical analysis. It should be quite possible,
however, to reach an approximate solution by numer-
ous direct microscopic measurements on thin sections
by transmitted light, and we hope to secure results of
this kind in the near future. In any case, the values
shown on this chart represent the extremes between
which the true values must lie, and we are of the opinion,
as intimated above, that the action during milling is
that the rubber phase will tend to become as evenly
distributed as possible, and that therefore the tetra-
hedral arrangement is the more nearly in accordance
with actual conditions.
The writer fully realizes that the foregoing analysis
hardly even scratches the surface of the problem of
the structure of compounded rubber. Of cardinal im-
portance are, for example, the direct measurement of
the surface tension between zinc oxide and rubber,
carbon blacks made .under different conditions and
rubber, and so on. When these values are once de-
termined the capacity factor of the surface energy as
measured by the average degree of dispersion of any
given pigment can in our opinion be most accurately
measured by its admixture under standard conditions
in a rubber compound, and the determination of the
decrease or increase in energy storage capacity as com-
pared with other samples of the same pigment. This
would seem to be of particular value in the case of the
finer pigments, such as the blacks, the individual par-
ticles of which are beyond the resolving power of our
microscopes.
Reverting to the title, "Rubber Energy," we see
that along with its already distracting array of prop-
erties chemical, rubber provides the thermodynamician
with plenty of nuts to crack. The interrelationships
of its thermal, mechanical, and surface energies make
up a field of research which has lain fallow long enough
and which should be zealously cultivated.
REACTIONS OF ACCELERATORS DURING VULCANIZA-
TION. II— A THEORY OF ACCELERATORS BASED
ON THE FORMATION OF POLYSULFIDES
DURING VULCANIZATION1
By Winfield Scott and C. W. Bedford
Goodvear Tire and Rubber Co., Akron, Ohio, and Quaker City
Rubber Co., Philadelphia, Pa.
The investigation of organic accelerators, as shown
by the literature of the past five or six years, appears
to be confined largely to a search for new compounds
or a combination of compounds to catalyze the ad-
dition of sulfur to rubber. It has been shown that
these accelerators are almost entirely organic nitrogen
compounds, and as a result nearly all classes of ni-
trogen-containing substances have been tried. Fur-
thermore, it has been shown that the nitrogen of such
compounds is basic or becomes basic during vulcani-
zation by the action of heat, sulfur, or hydrogen
sulfide.
It has been previously proposed that a sulfur re-
action of the accelerator is necessary, and certain re-
action products in some way make sulfur available
for vulcanization. In some cases a sulfur reaction is
doubtless necessary to form the true accelerator, which
is a polysulfide.
Ostromuislenski2 attributes the activation of sulfur
by aliphatic amines to the formation of thiozonides
of the type R-NH-S-S-S-NHR, which readily
■ Presented before the Rubber Division at the 60th Meeting of the
American Chemical Society, Chicago, III., September 6 to 10, 1920.
' Chem. Abs., 10 (1916), 1944.
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
give up their sulfur to the rubber. The formation of
thiozonides is illustrated by the following equation:
2R-NH2 + 4S
R-NH-S-S-S-NH-R + H2S
RNH2 + H:S
By this scheme, the true accelerator is produced to-
gether with hydrogen sulfide by the reaction of the
amine and sulfur. Such an explanation necessarily
excludes the tertiary amines, since they have no hy-
drogen attached to the nitrogen, and it is also limited
to those amines that react with sulfur at curing tem-
peratures. In the formation of thiozonides, hydrogen
sulfide is a by-product and does not function directl)
in producing a true accelerator.
Andre Dubosc1 states that a part of the curing
action of accelerators is due to the polymerizing effect
of thiocyanic acid produced by a sulfur reaction on the
accelerator. He illustrates these reactions by means
of equations, but makes no statement that such re-
action products were determined experimentally. As
an example of these reactions, it is stated that aniline
reacts with sulfur at 140°, in this manner:
C6H6NH2 + 4S >■ HCNS + 2HC=CH + CS, + H2S
The writers have been unable to duplicate these re-
sults, and no reference to any such reaction could be.
found in the literature on the subject. Dubosc at-
tributes the activation of sulfur entirely to the reac-
tion between hydrogen sulfide and sulfur dioxide.
It is known that vulcanization takes place if these
two gases are allowed to react in the presence of rub-
ber. Since the publication of the above-mentioned
article by Dubosc, a patent2 has been granted to
S. J. Peachey, covering the process. While there are
accelerators, such as />-nitrosodimethylaniline, which
generate both hydrogen sulfide and sulfur dioxide
during the cure, certainly the great majority of ac-
celerators do not activate sulfur in this way, since they
function in rubber stocks that are practically oxygen-
free.
The latest theory for the action of accelerators dur-
ing vulcanization is that of Kratz, Flower and Cool-
idgc.3 These writers attribute the accelerating action
of amines, such as aniline, to the formation of an un-
stable addition product of aniline and sulfur, in which
the sulfur is temporarily attached to the nitrogen,
making it pentavalent:
CtH[NH, + S
C6HSN
Z5
-H,
The compound thus formed gives up its sulfur to the
rubber and is then regenerated by a further reaction
with sulfur.
The writers believe that the mechanism of the ac-
tion of amines is represented differently from that
given by the above investigators, and that hydrogen
sulfide is one of the important factors in acceleration.
It is believed that, in general, amines catalyze the
addition of sulfur to rubber in the following manner:
• India Rubber World, 39 (1919), 5.
- Brit. Patent 129.826.
i This Journal, 12 (1920) 317.
SH
H
H
1
LNH2 + *S —
I
1
-> RNHj
1
SH
SH
Sx
As a specific example, dimethylamine, with hydrogen
sulfide and sulfur, forms a derivative of ammonium
polysulfide as follows:
(CH,)2NH + H,S > (CHii.XH.SH
(CH3)2NH2SH 4- xS > (CH:,12NH,SH
Polysulfide compounds similar to the above are con-
sidered to be the true accelerators that furnish the
sulfur necessary for vulcanization. That this type
of sulfur is available for vulcanization has been shown
by Ignaz Block,1 who states that hydrogen polysulfides
(H2S2 and H2S3) will cure rubber at ordinary tem-
peratures. C. O. Weber2 quotes Gerard and his work
showing that alkali polysulfides in concentrated solu-
tion will also vulcanize rubber.
ORGANIC ACCELERATORS
All organic accelerators do not function in the same
manner as the bases, and for this reason the writers
choose to divide accelerators into two classes.
I. Hydrogen Sulfide Polysulfide Accelerators — In this class be-
long those bases which form polysulfides similar to yellow
ammonium sulfide.
II. Carbo-sulfhydryl Polysulfide Accelerators — This includes
all accelerators that contain the grouping =C-SH, such as
the thioureas, dithiocarbamates, thiurams, mercaptans or the
disulfides which may be formed from them by oxidation or by
reaction with sulfur.3
To the first class belong all basic organic accelerators
or such compounds as produce basic accelerators un-
der curing conditions. Certain inorganic accelerators
may also be included. These will be discussed later
in the paper.
The second class also includes certain of the Schiff
bases4 which form thiourea derivatives by a sulfur
reaction during the cure. Further discussion of this
class will be reserved for a later paper. s
' D. R. P. 219.525.
."Chemistry of India Rubber," p. 47.
J Although the term polysulfide is applied to each elass of accelerators,
it should be noted that they are distinct types. In Class I. the polysulfide
sulfur is related to a sulfhydryl group attached to nitrogen, while in Class
II the polysulfide sulfur is held by a sulfhydryl group attached to carbon.
In the so-called disulfides and their higher sulfides, the hydrogen of the
sulfhydryl group is considered as having been eliminated in hydrogen sulfide.
« Bedford and Scott, This Journal. 12 (1920), 31 .
6 The reaction of carbon disulfide on amines to form thioureas and
hydrogen sulfide is reversible, and it is entirely possible that by the action
of hydrogen sulfide during vulcanization the thioureas are changed to the
more powerful dithiocarbamates which are intermediate to the complete
transformation to amine and carbon disulfide. It is also possible that the
thioureas may form polythio compounds direct, through the carbo-sulf-
hydryl gToup.
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
The phenylated guanidines belong to both classes,
since at curing temperatures they easily react with
hydrogen sulfide to form thioureas and free amines.
Diphenylguanidine, for example, gives thiocarbanilide
and ammonia.
The difference in behavior of the two above-men-
tioned classes of accelerators was well illustrated by
the following experiment: A rubber cement contain-
ing rubber, sulfur, and zinc oxide was divided into two
portions. To the first portion was added piperidyl
ammonium poly sulfide; there was no apparent change
after standing for 2 mo. To the second portion
was added an amount of piperidine equivalent to that
which was used in the first sample, and a small amount
of carbon disulfide was stirred into the mixture. This
cement jelled in less than 24 hrs., showing the well-
known higher curing power of the dithiocarbamates
as compared with basic amines and imines.
The present paper will deal with the first-mentioned
class of accelerators, i. e., with those accelerators which,
in the presence of hydrogen sulfide under curing con-
ditions, form polysulfides analogous to those of sodium
and ammonium.
The structural relationships of the polysulfides of
the nitrogen bases and the more positive metals are
not clearly understood at present, although it is known
that some of the sulfur is held in a more or less loose
form of chemical combination. This is evidenced by
the precipitation of sulfur from concentrated solutions
on dilution, and the generation of heat when sulfur
dissolves in sulfide or hydrosulfide solutions. It is
certain that the sulfur of polysulfides is quite different
from rhombic or a-sulfur, and that the aggregate
Sg is changed to the sulfur of polysulfides by the com-
bined action of hydrogen sulfide and basic accelerators.
It is a well-known fact that sulfur will react with
rubber resins and proteins at temperatures near 140°
with the formation of hydrogen sulfide. This hydrogen
sulfide in the presence of basic accelerators forms hy-
drosulfides which in turn take up sulfur to form poly-
sulfides. These polysulfides pass on part of their
sulfur to the rubber and constitute the true curing
agents. Such a mechanism applies also to the curing
action of alkali and alkaline-earth hydroxides. The
fact that basic magnesium carbonate will react with
hydrogen sulfide and sulfur in water suspension to
form polysulfide solutions no doubt accounts for its
mild accelerating power. Lime and magnesia do not
function well in deresinated rubbers where much of
the hydrogen sulfide producing materials have been
removed. The sulfides and polysulfides of the alkali
and alkaline-earth metals should function in deresin-
ated or synthetic rubbers.
The Bayer Company patent1 on basic organic acceler-
ators contains a broad claim covering all bases with a dis-
sociation constant greater than 1 X 10"8. This claim
covers those bases which readily react with hydrogen
sulfide and sulfur to form polysulfides at ordinary or
at curing temperatures. Weak bases such as aniline
cannot be expected to form polysulfides to the same
extent as strong bases like dimethylamine, since the
1 U. S. Patent 1,149.580.
formation of polysulfides is in some way dependent
upon basicity. It has been found that weak bases
such as aniline, />-toluidine, and quinoline, dissolve
more sulfur at 100° in the presence of hydrogen sulfide
than when it is absent. Aniline will dissolve about
1 per cent more sulfur at 100° and about 4 per cent
more at 130°.
The relative accelerating power of the organic bases
is dependent upon the facility with which they form
polysulfides and the extent to which they are able to
activate sulfur and make it available for the rubber.
This will, in some measure, be dependent upon the
basicity. In a previous paper by the writers1 it was
stated that at least a part of the accelerating action
of hexamethylenetetramine is due to the fact that
during the cure there are produced, among other
products, ammonia and carbon disulfide which, alone
or with basic products present in the rubber, form
dithiocarbamates. It may be added that "Hexn"
also forms hydrogen sulfide by sulfur reaction, which
with the ammonia undoubtedly forms ammonium
polysulfides. This accelerator may, therefore, be
classed under both types since it is both a hydrogen
sulfide and a carbo-sulfhydryl polysulfide accelerator.
Aldehyde ammonia, by the action of heat alone,
forms ammonia, while with sulfur it also gives hydrogen
sulfide. Heat also produces other bases such as the
alkyl pyridines or collidines. This material appears
to be solely a hydrogen sulfide polysulfide accelerator.
The ammonia condensation products of other aliphatic
aldehydes behave in a similar manner.
^-Phenylenediamine is an accelerator that is much
more active than would be assumed from its basicity.
At curing temperatures, this accelerator reacts with
sulfur to form large amounts of ammonia and hydrogen
sulfide together with certain weaker bases. If the
reaction be carried out under a cold reflux, the con-
denser will frequently become clogged with the white
solid compounds of ammonia and hydrogen sulfide
which are described by Roscoe and Schorlemmer.
The action of />-phenylenediamine in the cure is en-
tirely that of a hydrogen sulfide polysulfide accelerator.
The three above-mentioned accelerators are not
dependent on the rubber resins or proteins for their
supply of hydrogen sulfide, since this is one of their
sulfur reaction products. It is to be expected that
these accelerators will function in a deresinated or a
synthetic rubber, and the Bayer patents state that
this is true. It is also known that piperidine will cure
in a nitrogen-free rubber. Here we have a strong
base acting apparently without the aid of hydrogen
sulfide. Piperidine, however, reacts with sulfur at
temperatures lower than those used in vulcanization,
with the formation of hydrogen sulfide. Both the
sulfur reaction product and the unchanged piperidine
may then use this hydrogen sulfide to form polysulfides
with sulfur.
INORGANIC ACCELERATORS
Inorganic accelerators that function in the cure by
the removal of hydrogen sulfide the writers choose
to term "secondary accelerators," while those that
' hoe. cit.
128
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
function in the same manner as the organic polysulfide
accelerators may be classed with them as "primary
accelerators." A third class consists of those com-
pounds that are both primary and secondary accel-
erators.
I. Secondary Accelerators — Litharge, zinc oxide, etc., seem to
act no further than to form the corresponding sulfides, in con-
nection with hydrogen sulfide polysulfides.
II. Primary Accelerators — To this class belong the sulfides
and hydrosulfides of the alkali and alkaline-earth metals.
III. Accelerators That Are Both Primary and Secondary — In-
organic oxides and hydroxides function first as secondary accelera-
tors forming sulfides or hydrosulfides which then take up sulfur
and act as primary accelerators. Such accelerators are sodium
and calcium hydroxides, magnesium oxide and basic carbonate, etc.
Secondary accelerators are believed to function as
aids to organic polysulfides by breaking them up into
colloidal sulfur and the original nitrogen base. This
may be illustrated by the decolorization of polysulfide
solutions by litharge or zinc oxide. Ferric oxide does
not act as a secondary accelerator, and neither does
it readily decompose the polysulfide solutions. The
solubility of organic accelerators in sulfur and rubber
gives them much more intimate contact with hydrogen
sulfide at the time of its formation than is the case
with the comparatively large particles of litharge or
zinc oxide. Hydrogen sulfide is therefore available
for the formation of organic polysulfides before being
taken up by the. secondary accelerators. The de-
composition of a polysulfide by a secondary accelerator
regenerates the free base, which with more hydrogen
sulfide and sulfur re-forms the polysulfide. Secondary
accelerators do not act as true catalysts; once formed
into sulfides they do not react again with hydrogen
sulfide.
SUMMARY
1 — All organic accelerators are believed to function
through the formation of some type of polysulfide.
2 — Organic bases and compounds that form bases
during vulcanization are believed to form polysulfides
through the aid of hydrogen sulfide. These are termed
"hydrogen sulfide polysulfide accelerators."
3 — Thioureas, dithiocarbamates, thiurams, and
mercaptan compounds are believed to form polysul-
fides directly, or by first forming disulfides, and are
termed "carbo-sulfhydryl polysulfide accelerators."
4 — It is proposed that the function of such com-
pounds as litharge and zinc oxide may lie in the de-
composition of polysulfides into colloidal sulfur and
amines.
5 — Such inorganic compounds as sodium hydrox-
ide, calcium hydroxide and magnesium oxide are be-
lieved to function as "primary accelerators" through
the formation of inorganic polysulfides.
THE ACTION OF CERTAIN ORGANIC ACCELERATORS IN
THE VULCANIZATION OF RUBBER— III1
By G. D. Kratz, A. H. Flower and B. J. Shapiro
Falls Rubber Co., Cuyahoga Falls, Ohio
It has for some time been generally recognized that
although aniline is effective as an accelerator in the
1 Presented before the Rubber Division at the 60th Meeting of the
American Chemical Society, Chicago, 111., September 6 to 10, 1920.
absence of zinc oxide, diphenylthiourea functions but
mildly in the absence of, and strongly in the presence
of this substance. Reference to this effect has already
been made indirectly in the literature several times,
and recently Twiss1 has given curves for physical test
results which demonstrate quite clearly the effective-
ness of diphenylthiourea as an accelerator in the pres-
ence of zinc oxide. His statement that diphenyl-
thiourea is practically inert in the absence of zinc
oxide is, however, not in accord with our findings.
In a previous paper of this series2 we have shown
that in the acceleration of the vulcanization of a rubber-
sulfur mixture, the activity of one molecular part of
diphenylthiourea is less than that of an equimolecular
quantity of aniline, but equal to that of one molecular
part of aniline and one molecular part of phenyl mus-
tard oil.
Our former experiments, however, were confined to
the determination of sulfur coefficients at one cure
only. In the present instance, we desired to compare
the relative effects of aniline and diphenylthiourea
over a series of cures, and to effect this comparison
both by means of the sulfur coefficients and the physical
properties of the various mixtures and cures. Further,
it was desired to compare mixtures which contained
zinc oxide, as well as the rubber-sulfur mixtures
previously employed.
In the experimental part of this paper we have given
results obtained with six different mixtures, as follows
a rubber-sulfur control, a control which contained zinc
oxide, and similar mixtures which contained either one
molecular part of aniline or diphenylthiourea. All
of the mixtures were vulcanized for various intervals
over a wide range of time. After vulcanization, com-
parisons of sulfur coefficients and physical properties
were made.
Summarizing these results briefly, we found that, in
a rubber-sulfur mixture, the accelerating effect of
aniline is considerably greater than that of diphenyl-
thiourea, when judged" either by sulfur coefficients or
on the basis of the physical properties of the vulcanized
mixtures. In mixtures which contained zinc oxide,
however, the reverse was found to be true, and di-
phenylthiourea was more active than aniline when
judged by either of the above criteria. It was also
evident that in the case of the mixtures which con-
tained zinc oxide, although the tensile strength of the
mixture which was accelerated by diphenylthiourea
increased more rapidly than in the case of the mixture
accelerated by aniline, the same maximum tensile
strength was attained by each. The sulfur coefficients
at their respective maxima were practically identical.
While the maximum tensile strength of the rubber-
sulfur mixture which was accelerated by aniline was
the same as that obtained when zinc oxide was present
in the mixture, it was attained only at a much higher
sulfur coefficient. Lastly, it was also found that the
tensile strengths of the mixtures that contained zinc
oxide and which were accelerated by either aniline
or diphenylthiourea, particularly the latter, were in-
i J. Soc. Chem. Ind., 39 (1920), 1251.
' This Journal, 12 (1920), 317.
Feb., 1921
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY
129
creased tremendously during the first part of the vul-
canization, and at very low sulfur coefficients. This
would indicate the possibility of certain substances
(accelerators) increasing the physical properties of
a vulcanized mixture without greatly affecting the
sulfur coefficient.
This point is of interest as it already has been noted
by ourselves,1 Cranor,2 and others, that with mixtures
which contain zinc oxide and a strong organic acceler-
ator, the correct (or optimum) cure is obtained at
abnormally low sulfur coefficients when compared
with those obtained for unaccelerated mixtures. No
explanation has been offered for this phenomenon.
Bedford and Scott,3 however, regard diphenylthiourea
as the aniline salt of phenyldithiocarbamic acid after
H2S has been liberated. This salt is extremely un-
stable, owing to the weakly basic properties of aniline,
and in this respect, according to Krulla,4 is unlike the
metallic salts of the same acid. In this connection,
it is particularly pertinent to note that Bruni5 has
recently found the zinc salts of the mono and disub-
stituted dithiocarbamic acids to be violent accelerators.
It is quite possible, then, that such a salt may be formed
during the vulcanization process in mixtures which
contain both diphenylthiourea and zinc oxide;6 and
that, irrespective of its action as an accelerator, the
zinc portion of such a salt may be responsible for the
physical improvement imparted to the mixture.
Our present results, moreover, particularly when
interpreted with the assistance of the excess sulfur
coefficients obtained for the various mixtures at differ-
ent times of cure, show that when aniline is employed
as the accelerator in the presence of zinc oxide, the
effect of the latter substance is manifested almost en-
tirely in the physical properties of the mixture. When
aniline is replaced by diphenylthiourea the reverse is
true, and the activity of the original substance as an
accelerator is greatly increased when measured by
either the sulfur coefficients or physical properties.
In the latter instance, then, the zinc oxide most proba-
bly either assists in the decomposition of the diphenyl-
thiourea to a more active substance, or combines with
the decomposition or alteration products of the original
substance with the formation of a zinc salt, which is
responsible for the increase both in the sulfur coefficients
and tensile strength of the mixture. Our results with
aniline as the accelerator, however, do not indicate
the formation of such a salt.
Thus, in the presence of zinc oxide, the activity of
aniline and diphenylthiourea as accelerators appears
to be of a different nature. Evidently, an acid sub-
stance, probably a thiocarbamic acid, capable of re-
acting with zinc oxide, is formed as one of the de-
composition products of diphenylthiourea. The ex-
cess accelerating activity is attributed to this zinc salt.
' This Journal. 11 (1919), 30; Cliem. &■ Met. Eng., 20 (1919). 418.
'India Rubier World. 61 (1919), 137.
' This Journal, 12 (1920), 31.
' Ber., 46, 2669.
« Brit. Patents 140,387 and 140,388.
8 The action of diphenylthiourea with zinc oxide is apparently similar
to the action of the natural accelerator with magnesium oxide, as pointed
out in a previous paper (This Journal, 12 (1920), 971], In both cases the
oxide serves in a contributory capacity rather than as a primary accelerator.
It is obvious that no one oxide will activate all accelerators equally well
When aniline is employed as the accelerator, there is
no evidence of such salt formation.
EXPERIMENTAL PART
The present experiments were designed to effect a
comparison of the sulfur coefficients and physical
properties of representative mixtures when accelerated
by 0.01 gram-molecular quantities of either aniline or
diphenylthiourea. The six following mixtures were
employed for this purpose, and each was vulcanized
for a series of cures:
A — Rubber-sulfur control
B— Rubber, sulfur, and aniline
B-I — Rubber, sulfur, and diphenylthiourea
C — Rubber, sulfur, and zinc oxide control
D— Rubber, sulfur, zinc oxide, and aniline
D-I — Rubber, sulfur, zinc oxide, and diphenylthiourea
The quantities of each substance employed in these
mixtures are shown in Table I. The amounts of
Table I
Mix- Mix- Mix- Mix- Mix- Mix-
ture ture ture ture ture ture
Ingredient A B C D B I D-I
Rubber 100.00 100.00 100.00 100.00 100.00 100.00
Zinc oxide ... 100.00 100 00 ... 100.00
Sulfur 8.1 8.1 8.1 8.1 8.1 8.1
Aniline 0.93 ... 0.93
Diphenylthiourea ... ... ... 2.28 2.28
aniline or diphenylthiourea added to these respective
mixtures represent 0.01 gram-molecule of the acceler-
ator for each 100 g. of rubber in the mixture. Other-
wise, the same general method of procedure was adopted
in the course of this work as in that previously reported
in Part I.1
The rubber used was of good quality, first latex,
pale crepe, a different sample of the lot used in our
former experiments. The various mixtures were mixed
on the mill, vulcanized, and tested in the same manner
as before. The physical properties of the vulcanized
samples were determined on a Scott testing machine
of the vertical type, with the jaws opening at the rate
of 20 in. per min. A recovery period of 48 hrs. was
allowed before physical tests were made. Combined
sulfur was estimated by our method previously re-
ported in detail.2
The various mixtures were vulcanized at 141.5° C.
for different intervals of time up to 240 min.3 The
sulfur coefficients and physical properties of the dif-
ferent cures for each mixture were determined. These
results are given in detail in Table II and shown graph-
ically in Fig. 1. Generally speaking, the results ob-
tained were in good agreement, and fairly smooth
curves for physical properties were obtained.4
For brevity and clearness, the results obtained for
each mixture have been considered separately.
mixture a — This mixture of rubber and sulfur
served as a control only.
mixture b — Comparing Curves A and B, aniline
not only acts as an accelerator, but also slightly in-
creases the physical properties of a rubber-sulfur mix-
ture after vulcanization.
i This Journal, 12 (1920). 317.
'India Rubber World 61 (1920). 356.
1 In the experiments described in Parts I and II vulcanization was
carried on at a temperature of 148° C.
* Satisfactory physical test results for representation graphically are
obtainable with considerable difficulty. We have found it necessary, par.
ticularly when seeking results for stress-strain diagrams, to employ three
men, one to operate the machine and two to take readings.
li'.O
THE JOURNAL OF INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 13, No. 2
-Mixture A — n
Table II
: Vulcanized at 141.5°
. — Mixture B — .
-Mixture D — .
-—Mixture B-I^
, — Mixture D-I^
£-
-J
w
a*.
E
30 0.794 .. .. 1.126 545 1180 1.005
45 0.856 279 1250 1.317 1019 1170 1.055 306
60 1.038 .. .. 1.583 1228 1120 1.207 592
75 1.090 494 1220 1.898
90 1.531 709 1150 2.482 1621 1060 1.558 1041
120 2.089 871 1180 3.351 2046 1100 1.765 1815
150 2.236 1159 1130 4.033 2410 1100 2.237 1950
180 2.470 1521 1130 4.939 2670 1030 2.620 2032
210 3.179 1842 1100 5.264 2566 970 3.340 2184
240 3.751 2124 1060 6.268 2131 910 3.615 1978
1 Test pieces did not break.
mixture c — The inclusion of zinc oxide in Mixture
C was found to have little or no effect upon the sulfur co-
efficients when compared with the results obtained for A.
mixture r> — The sulfur coefficients obtained for this
mixture were found to be uniformly lower than the cor-
responding cures of B. Moreover, the maximum tensile
strength of D was attained at a much lower sulfur co-
efficient than in the case of B, although this maximum
tensile strength was almost the same in both instances.
30 60 90 /20 /SO /SO ZIO Z40
T/ME OF VULCANIZATION IN MINUTES
Fig. 1
mixture b-i — From the curves it is seen that in a
mixture of rubber and sulfur the activity of diphenyl-
thiourea is much less than that of aniline, when judged
by either sulfur coefficients or physical properties.
In fact, both the tensile strengths and final lengths
1.434
540
710
0.913
(')
1210
1.603
2210
8?0
680
1.490
1366
770
1.063
{*>
1360
1.912
2381
780
750
1.838
1819
770
1.335
<>1
1260
2.297
2442
790
1968
750
1,609
533
1230
2.623
760
2.382
2350
720
1.953
789
1230
2.962
2730
830
780
2.801
2808
780
2.496
1053
1210
3.755
2699
7 711
760
3.266
2721
770
3.109
1303
1150
4.521
2619
750
750
4.226
2663
740
4.027
1779
1110
5.357
2020
691)
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Books to Borrow
Open Library
TV News
Understanding 9/11